Device for sending point cloud data, method for sending point cloud data, device for receiving point cloud data, and method for receiving point cloud data

Abstract

According to an embodiment, a method for sending point cloud data may include the following steps: obtaining point cloud data; encoding geometric information of the positions of points including the point cloud data; generating one or more levels of detail (LODs) based on the geometric information, and selecting one or more neighboring points of each point to be attribute encoded based on the one or more LODs, wherein the one or more neighboring points of each selected point are within a maximum neighboring point distance; encoding attribute information of each point based on the one or more neighboring points of each selected point; and sending the encoded geometric information, the encoded attribute information, and signaling information.

Description

Technical Field

[0001] Embodiments relate to methods and apparatus for processing point cloud content. Background Art

[0002] Point cloud content is content represented by a point cloud, which is a collection of points belonging to a coordinate system representing a three-dimensional space. Point cloud content can represent media configured in three dimensions and is used to provide various services such as virtual reality (VR), augmented reality (AR), mixed reality (MR), XR (extended reality), and autonomous driving. However, tens of thousands to hundreds of thousands of point data are required to represent point cloud content. Therefore, methods for efficiently processing large amounts of point data are needed. Summary of the invention

[0003] Technical issues

[0004] The purpose of the present disclosure designed to solve the above-mentioned problems is to provide a point cloud data sending device, a point cloud data sending method, a point cloud data receiving device and a point cloud data receiving method for efficiently sending and receiving point clouds.

[0005] Another object of the present disclosure is to provide a point cloud data sending device, a point cloud data sending method, a point cloud data receiving device and a point cloud data receiving method for coping with waiting time and encoding / decoding complexity.

[0006] Another object of the present disclosure is to provide a point cloud data sending device, a point cloud data sending method, a point cloud data receiving device and a point cloud data receiving method for improving the compression performance of a point cloud by improving the encoding technology of attribute information of geometry-based point cloud compression (G-PCC).

[0007] Another object of the present disclosure is to provide a point cloud data sending device, a point cloud data sending method, a point cloud data receiving device and a point cloud data receiving method for enhancing compression efficiency while supporting parallel processing of attribute information of G-PCC.

[0008] Another object of the present disclosure is to provide a point cloud data sending device, a point cloud data sending method, a point cloud data receiving device and a point cloud data receiving method, which reduce the size of the attribute bit stream and enhance the compression efficiency of the attributes by selecting neighboring points for attribute prediction taking into account the attribute correlation between points in the content when the attribute information of G-PCC is encoded.

[0009] Another object of the present disclosure is to provide a point cloud data sending device, a point cloud data sending method, a point cloud data receiving device and a point cloud data receiving method that reduce the size of the attribute bit stream and enhance the compression efficiency of the attribute by applying the maximum neighboring point distance when the attribute information of G-PCC is encoded and selecting neighboring points for attribute prediction.

[0010] Another object of the present disclosure is to provide a point cloud data sending device, a point cloud data sending method, a point cloud data receiving device and a point cloud data receiving method that efficiently encode / decode various types of point cloud data by automatically calculating the maximum neighboring point range according to the density to obtain the maximum neighboring point distance when encoding the attribute information of the G-PCC.

[0011] The objects of the present disclosure are not limited to the objects mentioned above, and other objects of the present disclosure not mentioned above will become clear to those skilled in the art after reviewing the following description.

[0012] Technical Solution

[0013] To achieve these objectives and other advantages and in accordance with the purposes of the present disclosure, as implemented and broadly described herein, a point cloud data sending method may include the following steps: acquiring point cloud data; encoding geometric information of the positions of points including the point cloud data; generating one or more levels of detail (LODs) based on the geometric information, and selecting one or more neighboring points of each point to be attribute encoded based on the one or more LODs; encoding attribute information of each point based on the one or more neighboring points of each selected point; and sending the encoded geometric information, the encoded attribute information, and signaling information.

[0014] According to an embodiment, one or more neighboring points of each selected point may be located within a maximum neighboring point distance.

[0015] According to an embodiment, the maximum neighboring point distance may be determined based on the basic neighboring point distance and the maximum neighboring point range.

[0016] According to an embodiment, when the one or more LODs are generated based on an octree, the basic neighbor distance may be determined based on a diagonal distance of a node in a specific LOD.

[0017] According to an embodiment, selecting the one or more neighboring points may include the following steps: estimating the density of the point cloud data based on the basic neighboring point distance and the diagonal length of the bounding box of the point cloud data; and automatically calculating the maximum neighboring point range according to the estimated density.

[0018] According to an embodiment, information related to the maximum neighboring point range may be signaled in the signaling information.

[0019] In another aspect of the present disclosure, a point cloud data sending device may include: an acquirer configured to acquire point cloud data; a geometry encoder configured to encode geometric information of positions of points including the point cloud data; and an attribute encoder configured to generate one or more levels of detail (LODs) based on the geometric information, select one or more neighboring points of each point to be attribute encoded based on the one or more LODs, and encode attribute information of each point based on the one or more neighboring points of each selected point; and a transmitter configured to send the encoded geometry information, the encoded attribute information, and signaling information.

[0020] According to an embodiment, one or more neighboring points of each selected point may be located within a maximum neighboring point distance.

[0021] According to an embodiment, the maximum neighboring point distance may be determined based on the basic neighboring point distance and the maximum neighboring point range.

[0022] According to an embodiment, when the one or more LODs are generated based on an octree, the basic neighbor distance may be determined based on a diagonal distance of a node in a specific LOD.

[0023] The attribute encoder may estimate the density of the point cloud data based on the basic neighboring point distance and a diagonal length of a bounding box of the point cloud data, and automatically calculate the maximum neighboring point range according to the estimated density.

[0024] According to an embodiment, information related to the maximum neighboring point range may be signaled in the signaling information.

[0025] In another aspect of the present disclosure, a point cloud data receiving device may include: a receiver configured to receive geometric information, attribute information and signaling information; a geometry decoder configured to decode the geometric information based on the signaling information; an attribute decoder configured to generate one or more levels of detail (LODs) based on the geometric information, select one or more neighboring points of each point to be attribute encoded based on the one or more LODs, and decode the attribute information of each point based on the one or more neighboring points of each selected point and the signaling information; and a renderer configured to render point cloud data reconstructed based on the decoded geometric information and the decoded attribute information.

[0026] According to an embodiment, one or more neighboring points of each selected point may be located within a maximum neighboring point distance.

[0027] According to an embodiment, the maximum neighboring point distance may be determined based on the basic neighboring point distance and the maximum neighboring point range.

[0028] According to an embodiment, when the one or more LODs are generated based on an octree, the basic neighbor distance may be determined based on a diagonal distance of a node in a specific LOD.

[0029] The attribute decoder may estimate the density of the point cloud data based on the basic neighboring point distance and a diagonal length of a bounding box of the point cloud data, and automatically calculate the maximum neighboring point range according to the estimated density.

[0030] According to an embodiment, the attribute decoder may acquire the maximum neighboring point range from the signaling information.

[0031] Beneficial Effects

[0032] The point cloud data sending method, point cloud data sending device, point cloud data receiving method and point cloud receiving device according to the embodiments can provide high-quality point cloud services.

[0033] According to the point cloud data sending method, point cloud data sending device, point cloud data receiving method and point cloud receiving device of the implementation method, various video encoding and decoding methods can be implemented.

[0034] The point cloud data transmitting method, the point cloud data transmitting device, the point cloud data receiving method, and the point cloud receiving device according to the embodiments can provide general point cloud content such as an automatic driving service (or autonomous driving service).

[0035] The point cloud data sending method, point cloud data sending device, point cloud data receiving method and point cloud data receiving device according to the embodiments can perform spatial adaptive segmentation of point cloud data to independently encode and decode point cloud data, thereby improving parallel processing and providing scalability.

[0036] According to the implementation manner, the point cloud data sending method, the point cloud data sending device, the point cloud data receiving method and the point cloud data receiving device can perform encoding and decoding by spatially dividing the point cloud data in units of tiles and / or slices and thereby signaling necessary data thereof, thereby improving the encoding and decoding performance of the point cloud.

[0037] The point cloud data sending method and apparatus and the point cloud data receiving method and apparatus according to the embodiment reduce the size of the attribute bitstream and enhance the compression efficiency of the attribute by selecting neighboring points for attribute prediction in consideration of the attribute correlation between points of the content when the attribute information of the G-PCC is encoded.

[0038] According to an embodiment, a point cloud data sending method, a point cloud data sending device, a point cloud data receiving method and a point cloud data receiving device can cause neighboring points for attribute prediction to be selected based on the maximum neighboring point distance when encoding attribute information of G-PCC, thereby reducing the size of the attribute bit stream and improving the attribute compression efficiency.

[0039] According to an embodiment, a point cloud data transmitting method, a point cloud data transmitting device, a point cloud data receiving method, and a point cloud data receiving device may cause the neighboring points for attribute prediction to be selected based on the maximum neighboring point distance when encoding the attribute information of the G-PCC, thereby removing points outside the maximum neighboring point distance, that is, points in the neighboring point set that hinder prediction. Therefore, when the correlation of the attribute is low due to the attribute characteristics of the content, the predicted average value may be increased so that the residual value with the point attribute value does not increase, and weights may not be assigned to unimportant information. Therefore, the attribute bitstream size may be reduced and the PSNR accuracy may be increased.

[0040] According to the implementation, the point cloud data sending method, point cloud data sending device, point cloud data receiving method and point cloud data receiving device can configure the maximum neighboring point range according to the LOD configuration method, service characteristics or content characteristics. Accordingly, a maximum neighboring point range suitable for attribute characteristics can be provided.

[0041] According to an embodiment, a point cloud data transmitting method, a point cloud data transmitting device, a point cloud data receiving method, and a point cloud data receiving device can configure the maximum range of neighboring points according to an LOD configuration method, service characteristics, or content characteristics. Accordingly, various types of point cloud data can be efficiently encoded / decoded.

[0042] According to an embodiment, a point cloud data transmitting method, a point cloud data transmitting device, a point cloud data receiving method, and a point cloud data receiving device may automatically calculate a maximum neighboring point range according to density, thereby providing optimal compression efficiency.

[0043] According to an embodiment, a point cloud data transmission method, a point cloud data transmission device, a point cloud data receiving method, and a point cloud data receiving device can estimate the maximum neighboring point distance using at least one of an octree-based calculation of the maximum neighboring point distance, a distance-based calculation of the maximum neighboring point distance, a sampling-based calculation of the maximum neighboring point distance, a calculation of the maximum neighboring point distance by calculating the average difference of Morton codes between LODs, or a calculation of the maximum neighboring point distance by calculating the average distance difference between LODs, and signal relevant information. Thus, the attribute compression / decompression efficiency can be improved by an optimal set of neighboring points. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The accompanying drawings are included to provide a further understanding of the present disclosure and are incorporated in and constitute a part of this application. The accompanying drawings illustrate embodiments of the present disclosure and together with the description serve to explain the principles of the present disclosure. In the drawings:

[0045] Figure 1 An exemplary point cloud content providing system according to an embodiment is illustrated.

[0046] Figure 2 is a block diagram illustrating a point cloud content providing operation according to an embodiment.

[0047] Figure 3 An exemplary process of capturing point cloud video according to an embodiment is illustrated.

[0048] Figure 4 An exemplary block diagram of a point cloud video encoder according to an embodiment is illustrated.

[0049] Figure 5 An example of voxels in a 3D space according to an embodiment is illustrated.

[0050] Figure 6 Examples of octrees and occupancy codes according to embodiments are illustrated.

[0051] Figure 7 An example of a neighboring node pattern according to an embodiment is illustrated.

[0052] Figure 8 An example of point configuration of point cloud content for each LOD according to an embodiment is illustrated.

[0053] Fig. 9 An example of point configuration of point cloud content for each LOD according to an embodiment is illustrated.

[0054] Fig.10 An example of a block diagram of a point cloud video decoder according to an embodiment is illustrated.

[0055] Fig.11 An example of a point cloud video decoder according to an embodiment is illustrated.

[0056] Fig.12 A configuration of point cloud video encoding of a transmitting device according to an embodiment is illustrated.

[0057] Fig.13 A configuration of point cloud video decoding of a receiving device according to an embodiment is illustrated.

[0058] Fig.14 An exemplary structure operatively connectable with the method / apparatus for transmitting and receiving point cloud data according to an embodiment is illustrated.

[0059] Fig.15 An example of a point cloud transmitting device according to an embodiment is illustrated.

[0060] Fig.16 (a) to Fig.16 (c) illustrates an implementation of segmenting the bounding box into one or more tiles.

[0061] Fig.17 Examples of a geometry encoder and an attribute encoder according to an embodiment are illustrated.

[0062] Fig.18 is a diagram illustrating an example of generating LOD based on an octree according to an embodiment.

[0063] Fig.19 is a diagram illustrating an example of arranging points in a point cloud in a Morton code order according to an embodiment.

[0064] Fig. 20 is a diagram illustrating an example of searching for neighboring points based on LOD according to an embodiment.

[0065] Fig.21 (a) and Fig.21 (b) illustrates an example of point cloud content according to an embodiment.

[0066] Fig. 22 (a) and Fig. 22 (b) illustrates an example of an average distance, a minimum distance, and a maximum distance of each point belonging to a neighboring point set according to an embodiment.

[0067] Fig.23 (a) illustrates an example of obtaining a diagonal distance of an octree node of LOD0 according to an embodiment.

[0068] Fig.23 (b) illustrates an example of obtaining the diagonal distance of the octree node of LOD1 according to an embodiment.

[0069] Fig.24 (a) and Fig.24 (b) illustrates an example of a range that may be selected as neighboring points at each LOD.

[0070] Fig.25 An example of a basic neighboring point distance belonging to each LOD according to an embodiment is illustrated.

[0071] Fig.26 is a flowchart illustrating automatic calculation of a maximum neighboring point range based on density according to an embodiment of the present disclosure.

[0072] Fig. 27is a flowchart illustrating automatic calculation of a maximum neighboring point range based on density according to another embodiment of the present disclosure.

[0073] Fig.28 Another example of searching for neighboring points based on LOD according to an embodiment is illustrated.

[0074] Fig.29 Another example of searching for neighboring points based on LOD according to an embodiment is illustrated.

[0075] Fig.30 An example of a point cloud receiving device according to an embodiment is illustrated.

[0076] Fig.31 Examples of a geometry decoder and an attribute decoder according to an embodiment are illustrated.

[0077] Fig.32 An exemplary bitstream structure for transmitting / receiving point cloud data according to an embodiment is illustrated.

[0078] Fig.33 An exemplary bitstream structure of point cloud data according to an embodiment is illustrated.

[0079] Fig.34 The connection relationship between components in a bit stream of point cloud data according to an embodiment is illustrated.

[0080] Fig.35 An embodiment of a syntax structure of a sequence parameter set according to an embodiment is illustrated.

[0081] Fig.36 An implementation of a syntax structure of a geometry parameter set according to an implementation is illustrated.

[0082] Fig.37 An embodiment of a syntax structure of a property parameter set according to an embodiment is illustrated.

[0083] Fig.38 An embodiment of a syntax structure of a tile parameter set according to an embodiment is illustrated.

[0084] Fig.39 An implementation of the syntax structure of a geometry slice bitstream() according to an implementation is illustrated.

[0085] Fig.40 An embodiment of a syntax structure of a geometry slice header according to an embodiment is illustrated.

[0086] Fig.41 An embodiment of a syntax structure of geometry slice data according to an embodiment is illustrated.

[0087] Fig.42An embodiment of a syntax structure of an attribute slice bitstream() according to an embodiment is illustrated.

[0088] Fig.43 An embodiment of a syntax structure of an attribute slice header according to an embodiment is illustrated.

[0089] Fig.44 An embodiment of a syntax structure of attribute slice data according to an embodiment is illustrated.

[0090] Fig.45 is a flow chart of a method of transmitting point cloud data according to an embodiment.

[0091] Fig.46 is a flowchart of a method of receiving point cloud data according to an embodiment. DETAILED DESCRIPTION

[0092] Now, a description will be given in detail according to the exemplary embodiments disclosed herein with reference to the accompanying drawings. For the purpose of brief description with reference to the accompanying drawings, the same reference numerals may be provided for the same or equivalent components, and their descriptions will not be repeated. It should be noted that the following examples are only used to embody the present disclosure and do not limit the scope of the present disclosure. What can be easily inferred from the detailed description and examples of the present disclosure by experts in the technical field to which the present invention belongs will be interpreted as being within the scope of the present disclosure.

[0093] The detailed description in this specification should be interpreted in all aspects as illustrative rather than restrictive. The scope of the present disclosure should be determined by the appended claims and their legal equivalents, and all changes coming within the meaning and equivalency range of the appended claims are intended to be embraced herein.

[0094] Now, reference will be made in detail to the preferred embodiments of the present disclosure, examples of which are illustrated in the accompanying drawings. The detailed description given below with reference to the accompanying drawings is intended to explain the exemplary embodiments of the present disclosure, rather than to show the only embodiments that can be implemented according to the present disclosure. The following detailed description includes specific details to provide a thorough understanding of the present disclosure. However, it will be apparent to those skilled in the art that the present disclosure can be practiced without these specific details. Although most of the terms used in this specification have been selected from the general terms widely used in the art, the applicant has arbitrarily selected some terms, and their meanings will be explained in detail as needed in the following description. Therefore, the present disclosure should be understood based on the original meaning of the terms rather than their simple names or meanings. In addition, the following drawings and detailed descriptions should not be interpreted as being limited to the embodiments specifically described, but should be interpreted as including equivalents or substitutes of the embodiments described in the drawings and detailed descriptions.

[0095] Figure 1An exemplary point cloud content providing system according to an embodiment is shown.

[0096] Figure 1 The point cloud content providing system illustrated in the example may include a transmitting device 10000 and a receiving device 10004. The transmitting device 10000 and the receiving device 10004 can perform wired or wireless communication to transmit and receive point cloud data.

[0097] The point cloud data sending device 10000 according to an embodiment can protect and process the point cloud video (or point cloud content), and send the point cloud video (or point cloud content). According to an embodiment, the sending device 10000 may include a fixed station, a base station transceiver system (BTS), a network, an artificial intelligence (AI) device and / or system, a robot, an AR / VR / XR device and / or a server. According to an embodiment, the sending device 10000 may include a device configured to communicate with a base station and / or other wireless devices using a radio access technology (e.g., 5G new RAT (NR), long term evolution (LTE)), a robot, a vehicle, an AR / VR / XR device, a portable device, a home appliance, an Internet of Things (IoT) device, and an AI device / server.

[0098] According to the implementation manner, the sending device 10000 includes a point cloud video acquisition unit 10001, a point cloud video encoder 10002 and / or a transmitter (or communication module) 10003.

[0099] The point cloud video acquisition unit 10001 according to the embodiment acquires the point cloud video through a processing process such as capturing, synthesizing or generating. The point cloud video is a point cloud content represented by a point cloud as a collection of points in a 3D space, and can be referred to as point cloud video data. The point cloud video according to the embodiment may include one or more frames. A frame represents a still image / picture. Therefore, the point cloud video may include a point cloud image / frame / picture, and may be referred to as a point cloud image, frame or picture.

[0100] The point cloud video encoder 10002 according to the implementation manner encodes the acquired point cloud video data. The point cloud video encoder 10002 may encode the point cloud video data based on point cloud compression coding. The point cloud compression coding according to the implementation manner may include geometry-based point cloud compression (G-PCC) coding and / or video-based point cloud compression (V-PCC) coding or next generation coding. The point cloud compression coding according to the implementation manner is not limited to the above-mentioned implementation manner. The point cloud video encoder 10002 may output a bit stream containing the encoded point cloud video data. The bit stream may contain not only the encoded point cloud video data, but also signaling information related to the encoding of the point cloud video data.

[0101] The transmitter 10003 according to the embodiment transmits a bitstream containing the encoded point cloud video data. The bitstream according to the embodiment is encapsulated in a file or segment (e.g., a streaming segment) and transmitted through various networks such as a broadcast network and / or a broadband network. Although not shown in the figure, the transmitting device 10000 may include a packager (or a package module) configured to perform a packaging operation. According to the embodiment, the packager may be included in the transmitter 10003. According to the embodiment, the file or segment may be sent to the receiving device 10004 through a network, or may be stored in a digital storage medium (e.g., USB, SD, CD, DVD, Blu-ray, HDD, SSD, etc.). The transmitter 10003 according to the embodiment can perform wired / wireless communication with the receiving device 10004 (or the receiver 10005) through a network such as 4G, 5G, 6G, etc. In addition, the transmitter may perform necessary data processing operations according to a network system (e.g., a 4G, 5G or 6G communication network system). The transmitting device 10000 may send the encapsulated data in an on-demand manner.

[0102] The receiving device 10004 according to an embodiment includes a receiver 10005, a point cloud video decoder 10006 and / or a renderer 10007. According to an embodiment, the receiving device 10004 may include a device configured to communicate with a base station and / or other wireless devices using a radio access technology (e.g., 5G New RAT (NR), Long Term Evolution (LTE)), a robot, a vehicle, an AR / VR / XR device, a portable device, a home appliance, an Internet of Things (IoT) device, and an AI device / server.

[0103] The receiver 10005 according to the embodiment receives a bitstream containing point cloud video data or a file / segment in which the bitstream is encapsulated from a network or storage medium. The receiver 10005 can perform necessary data processing according to a network system (e.g., a communication network system such as 4G, 5G, 6G, etc.). The receiver 10005 according to the embodiment can decapsulate the received file / segment and output the bitstream. According to the embodiment, the receiver 10005 may include a decapsulator (or decapsulation module) configured to perform a decapsulation operation. The decapsulator can be implemented as an element (or component or module) separate from the receiver 10005.

[0104] The point cloud video decoder 10006 decodes the bitstream containing the point cloud video data. The point cloud video decoder 10006 can decode the point cloud video data according to the method of encoding the point cloud video data (for example, in the inverse process of the operation of the point cloud video encoder 10002). Therefore, the point cloud video decoder 10006 can decode the point cloud video data by performing point cloud decompression encoding, which is the inverse process of point cloud compression. Point cloud decompression encoding includes G-PCC encoding.

[0105] The renderer 10007 renders the decoded point cloud video data. The renderer 10007 may output the point cloud content by rendering not only the point cloud video data but also the audio data. According to an embodiment, the renderer 10007 may include a display configured to display the point cloud content. According to an embodiment, the display may be implemented as a separate device or component rather than being included in the renderer 10007.

[0106] The arrow indicated by the dotted line in the figure represents the transmission path of the feedback information obtained by the receiving device 10004. Feedback information is information used to reflect the interactivity with the user consuming the point cloud content, and includes information about the user (e.g., head orientation information, viewport information, etc.). In particular, when the point cloud content is the content of a service that requires interaction with the user (e.g., an autonomous driving service, etc.), the feedback information may be provided to the content sender (e.g., the sending device 10000) and / or the service provider. Depending on the embodiment, the feedback information may be used in the receiving device 10004 and the sending device 10000, or may not be provided.

[0107] The head orientation information according to the embodiment is information about the user's head position, orientation, angle, movement, etc. The receiving device 10004 according to the embodiment can calculate the viewport information based on the head orientation information. The viewport information can be information about the area of ​​the point cloud video that the user is watching. The viewpoint is the point through which the user is watching the point cloud video, and can refer to the center point of the viewport area. That is, the viewport is an area centered on the viewpoint, and the size and shape of the area can be determined by the field of view (FOV). Therefore, in addition to the head orientation information, the receiving device 10004 can also extract the viewport information based on the vertical or horizontal FOV supported by the device. In addition, the receiving device 10004 performs gaze analysis, etc. to check the way the user consumes the point cloud, the area where the user gazes in the point cloud video, the gaze time, etc. According to the embodiment, the receiving device 10004 can send feedback information including the gaze analysis result to the sending device 10000. The feedback information according to the embodiment can be obtained in the rendering and / or display process. The feedback information according to an embodiment may be protected by one or more sensors included in the receiving device 10004. According to an embodiment, the feedback information may be protected by the renderer 10007 or a separate external element (or device, component, etc.).

[0108] Figure 1 The dotted line in represents the process of sending feedback information protected by the renderer 10007. The point cloud content providing system can process (encode / decode) the point cloud data based on the feedback information. Therefore, the point cloud video decoder 10006 can perform a decoding operation based on the feedback information. The receiving device 10004 can send the feedback information to the sending device 10000. The sending device 10000 (or the point cloud video encoder 10002) can perform an encoding operation based on the feedback information. Therefore, the point cloud content providing system can efficiently process necessary data (for example, point cloud data corresponding to the user's head position) based on the feedback information instead of processing (encoding / decoding) the entire point cloud data, and provide point cloud content to the user.

[0109] According to an embodiment, the transmitting device 10000 may be referred to as an encoder, a transmitting device, a transmitter, a transmitting system, etc., and the receiving device 10004 may be referred to as a decoder, a receiving device, a receiver, a receiving system, etc.

[0110] (Through a series of processes of obtaining / encoding / sending / decoding / rendering) according to the implementation Figure 1 The point cloud data processed in the point cloud content providing system may be referred to as point cloud content data or point cloud video data. According to an embodiment, point cloud content data may be used as a concept covering metadata or signaling information related to point cloud data.

[0111] Figure 1The elements of the point cloud content providing system illustrated in the example may be implemented by hardware, software, a processor and / or a combination thereof.

[0112] Figure 2 is a block diagram illustrating a point cloud content providing operation according to an embodiment.

[0113] Figure 2 The block diagram shows Figure 1 The operation of the point cloud content providing system described in . As described above, the point cloud content providing system can process point cloud data based on point cloud compression coding (e.g., G-PCC).

[0114] According to the embodiment, the point cloud content providing system (e.g., the point cloud sending device 10000 or the point cloud video acquisition unit 10001) can acquire a point cloud video (20000). The point cloud video is represented by a point cloud belonging to a coordinate system for representing a 3D space. According to the embodiment, the point cloud video may include a Ply (polygon file format or Stanford triangle format) file. When the point cloud video has one or more frames, the acquired point cloud video may include one or more Ply files. The Ply file contains point cloud data such as point geometry and / or attributes. The geometry includes the position of the point. The position of each point can be represented by a parameter (e.g., the value of the X, Y and Z axes) representing a three-dimensional coordinate system (e.g., a coordinate system consisting of X, Y and Z axes). Attributes include attributes of the point (e.g., information about the texture, color (YCbCr or RGB), reflectivity r, transparency, etc. of each point). A point has one or more attributes. For example, a point may have an attribute as a color or two attributes as color and reflectivity.

[0115] Depending on the implementation, the geometry may be referred to as a position, geometry information, geometry data, etc., and the attributes may be referred to as attributes, attribute information, attribute data, etc.

[0116] The point cloud content providing system (for example, the point cloud sending device 10000 or the point cloud video acquiring unit 10001) can obtain point cloud data based on information related to the acquisition and processing of the point cloud video (for example, depth information, color information, etc.).

[0117] According to an embodiment, a point cloud content providing system (e.g., a transmitting device 10000 or a point cloud video encoder 10002) can perform encoding processing on point cloud data (20001). The point cloud content providing system can encode the point cloud data based on point cloud compression coding. As described above, the point cloud data may include the geometric structure and attributes of the points. Therefore, the point cloud content providing system can perform geometric coding for encoding the geometric structure and output a geometric bitstream. The point cloud content providing system can perform attribute coding for encoding the attributes and output an attribute bitstream. According to an embodiment, the point cloud content providing system can perform attribute coding based on geometric coding. According to an embodiment, the geometric bitstream and the attribute bitstream can be multiplexed and output as a bitstream. According to an embodiment, the bitstream may also include signaling information related to geometric coding and attribute coding.

[0118] According to the embodiment, the point cloud content providing system (eg, the sending device 10000 or the sender 10003) may send the encoded point cloud data. Figure 1 As illustrated in , the encoded point cloud data can be represented by a geometry bitstream and an attribute bitstream. In addition, the encoded point cloud data can be sent in the form of a bitstream together with signaling information related to the encoding of the point cloud data (e.g., signaling information related to geometry encoding and attribute encoding). The point cloud content providing system can encapsulate a bitstream carrying the encoded point cloud data and send the bitstream in the form of a file or a segment.

[0119] According to the embodiment, the point cloud content providing system (e.g., the receiving device 10004 or the receiver 10005) can receive a bit stream containing the encoded point cloud data. In addition, the point cloud content providing system (e.g., the receiving device 10004 or the receiver 10005) can demultiplex the bit stream.

[0120] The point cloud content providing system (e.g., receiving device 10004 or point cloud video decoder 10006) can decode the encoded point cloud data (e.g., geometry bitstream, attribute bitstream) sent in the bitstream. The point cloud content providing system (e.g., receiving device 10004 or point cloud video decoder 10006) can decode the point cloud video data based on the signaling information related to the encoding of the point cloud video data contained in the bitstream. The point cloud content providing system (e.g., receiving device 10004 or point cloud video decoder 10006) can decode the geometry bitstream to reconstruct the position (geometry) of the point. The point cloud content providing system can reconstruct the attributes of the point by decoding the attribute bitstream based on the reconstructed geometry. The point cloud content providing system (e.g., receiving device 10004 or point cloud video decoder 10006) can reconstruct the point cloud video based on the position according to the reconstructed geometry and the decoded attributes.

[0121] According to the implementation, the point cloud content providing system (e.g., receiving device 10004 or renderer 10007) can render the decoded point cloud data. The point cloud content providing system (e.g., receiving device 10004 or renderer 10007) can use various rendering methods to render the geometric structure and attributes decoded by the decoding process. The points in the point cloud content can be rendered as vertices with a certain thickness, a cube with a specific minimum size centered on the corresponding vertex position, or a circle centered on the corresponding vertex position. All or part of the rendered point cloud content is provided to the user through a display (e.g., a VR / AR display, a common display, etc.).

[0122] The point cloud content providing system (e.g., receiving device 10004) according to the embodiment may protect the feedback information (20005). The point cloud content providing system may encode and / or decode the point cloud data based on the feedback information. Figure 1 The feedback information and operations described are the same, so a detailed description thereof is omitted.

[0123] Figure 3 An exemplary process of capturing point cloud video according to an embodiment is illustrated.

[0124] Figure 3 Reference Figure 1 to Figure 2 An exemplary point cloud video capture process for a point cloud content providing system is described.

[0125] Point cloud content includes point cloud videos (images and / or videos) representing objects and / or environments located in various 3D spaces (e.g., 3D spaces representing real environments, 3D spaces representing virtual environments, etc.). Therefore, a point cloud content providing system according to an embodiment may use one or more cameras (e.g., an infrared camera capable of protecting depth information, an RGB camera capable of extracting color information corresponding to the depth information, etc.), a projector (e.g., an infrared pattern projector for protecting depth information), LiDRA, etc. to capture point cloud videos. A point cloud content providing system according to an embodiment may extract the shape of a geometric structure composed of points in a 3D space from the depth information, and extract the attributes of each point from the color information to protect the point cloud data. Images and / or videos according to an embodiment may be captured based on at least one of an inward-facing technology and an outward-facing technology.

[0126] Figure 3The left portion of exemplifies an inward-facing technique. An inward-facing technique refers to a technique of capturing an image of a central object with one or more cameras (or camera sensors) disposed around the central object. The inward-facing technique can be used to generate point cloud content that provides a 360-degree image of a key object to a user (e.g., VR / AR content that provides a 360-degree image of an object (e.g., a key object such as a character, player, object, or actor) to a user).

[0127] Figure 3 The right portion of exemplifies an outward facing technique. An outward facing technique refers to a technique in which the environment of a central object rather than an image of the central object is captured using one or more cameras (or camera sensors) disposed around the central object. Point cloud content for providing a surrounding environment as it appears from the user's perspective (e.g., content representing an external environment that can be provided to a user of an autonomous vehicle) can be generated using an outward facing technique.

[0128] like Figure 3 As shown in , point cloud content can be generated based on the capture operation of one or more cameras. In this case, the coordinate system is different in each camera, so the point cloud content providing system can calibrate one or more cameras to set the global coordinate system before the capture operation. In addition, the point cloud content providing system can generate point cloud content by synthesizing any image and / or video with the image and / or video captured by the above-mentioned capture technology. The point cloud content providing system may not perform when it generates point cloud content representing a virtual space. Figure 3 The point cloud content providing system according to the embodiment may perform post-processing on the captured image and / or video. In other words, the point cloud content providing system may remove unnecessary areas (e.g., background), identify the space to which the captured image and / or video is connected, and perform an operation of filling the space hole when there is a space hole.

[0129] The point cloud content providing system can generate a piece of point cloud content by performing coordinate transformation on the points of the point cloud video secured from each camera. The point cloud content providing system can perform coordinate transformation on the points based on the coordinates of each camera position. Therefore, the point cloud content providing system can generate content representing a wide range, or can generate point cloud content with high density of points.

[0130] Figure 4 An exemplary point cloud video encoder according to an embodiment is illustrated.

[0131] Figure 4 Shows Figure 1An example of a point cloud video encoder 10002. The point cloud video encoder reconstructs and encodes point cloud data (e.g., the location and / or attributes of points) to adjust the quality of the point cloud content (e.g., lossless, lossy, or near-lossless) according to network conditions or applications. When the total size of the point cloud content is large (e.g., for 30fps, giving 60Gbps of point cloud content), the point cloud content providing system may not be able to stream the content in real time. Therefore, the point cloud content providing system can reconstruct the point cloud content based on the maximum target bit rate to provide the point cloud content according to the network environment, etc.

[0132] As reference Figure 1 to Figure 2 As described above, the point cloud video encoder can perform geometry encoding and attribute encoding. Geometry encoding is performed before attribute encoding.

[0133] According to an embodiment, the point cloud video encoder includes a coordinate transformation unit 40000, a quantization unit 40001, an octree analysis unit 40002, a surface approximation analysis unit 40003, an arithmetic encoder 40004, a geometric reconstruction unit 40005, a color transformation unit 40006, an attribute transformation unit 40007, a RAHT unit 40008, an LOD generation unit 40009, a lifting transformation unit 40010, a coefficient quantization unit 40011 and / or an arithmetic encoder 40012.

[0134] The coordinate transformation unit 40000, the quantization unit 40001, the octree analysis unit 40002, the surface approximation analysis unit 40003, the arithmetic encoder 40004 and the geometric reconstruction unit 40005 can perform geometric coding. The geometric coding according to the embodiment may include octree geometric coding, direct coding, trisoup (trisoup) geometric coding and entropy coding. Direct coding and trisoup geometric coding are applied selectively or in combination. Geometric coding is not limited to the above examples.

[0135] As shown in the figure, the coordinate transformation unit 40000 according to the embodiment receives the position and transforms it into coordinates. For example, the position can be transformed into position information in a three-dimensional space (e.g., a three-dimensional space represented by an XYZ coordinate system). The position information in the three-dimensional space according to the embodiment can be referred to as geometric information.

[0136] The quantization unit 40001 according to the embodiment quantizes the geometric information. For example, the quantization unit 40001 can quantize the points based on the minimum position value of all points (e.g., the minimum value on each of the X, Y, and Z axes). The quantization unit 40001 performs the following quantization operation: the difference between the position value of each point and the minimum position value is multiplied by a preset quantization scaling value, and then the nearest integer value is found by rounding the value obtained by the multiplication. Therefore, one or more points can have the same quantized position (or position value). The quantization unit 40001 according to the embodiment performs voxelization based on the quantized position to reconstruct the quantized point. Voxelization means the smallest unit representing position information in 3D space. The points of the point cloud content (or 3D point cloud video) according to the embodiment can be included in one or more voxels. The term voxel, which is a compound word of volume and pixel, refers to a 3D cubic space generated when the 3D space is divided into units (unit = 1.0) based on the axes representing the 3D space (e.g., X-axis, Y-axis, and Z-axis). The quantization unit 40001 may match a group of points in a 3D space with a voxel. According to an embodiment, a voxel may include only one point. According to an embodiment, a voxel may include one or more points. In order to represent a voxel as a point, the position of the center point of the voxel may be set based on the position of one or more points included in the voxel. In this case, the properties of all positions included in a voxel may be combined and assigned to the voxel.

[0137] The octree analysis unit 40002 according to the embodiment performs octree geometry encoding (or octree encoding) to present voxels in an octree structure. The octree structure represents points matched with voxels based on the octree structure.

[0138] The surface approximation analysis unit 40003 according to the embodiment may analyze and approximate the octree. The octree analysis and approximation according to the embodiment is a process of analyzing a region including a plurality of points to efficiently provide an octree and voxelization.

[0139] The arithmetic encoder 40004 according to the embodiment performs entropy coding on the octree and / or the approximate octree. For example, the coding scheme includes arithmetic coding. As a result of the coding, a geometry bitstream is generated.

[0140] The color transform unit 40006, the attribute transform unit 40007, the RAHT unit 40008, the LOD generation unit 40009, the lifting transform unit 40010, the coefficient quantization unit 40011 and / or the arithmetic encoder 40012 perform attribute coding. As described above, a point may have one or more attributes. The attribute coding according to the embodiment is also applied to the attributes of a point. However, when the attribute (e.g., color) includes one or more elements, the attribute coding is applied independently to each element. The attribute coding according to the embodiment includes color transform coding, attribute transform coding, regional adaptive hierarchical transform (RAHT) coding, interpolation-based hierarchical nearest neighbor prediction (prediction transform) coding, and interpolation-based hierarchical nearest neighbor prediction coding with an update / lifting step (lifting transform). Depending on the point cloud content, the above-mentioned RAHT coding, prediction transform coding, and lifting transform coding can be selectively used, or a combination of one or more coding schemes can be used. The attribute coding according to the embodiment is not limited to the above examples.

[0141] The color conversion unit 40006 according to an embodiment performs color conversion encoding of the color value (or texture) included in the conversion attribute. For example, the color conversion unit 40006 can convert the format of the color information (e.g., from RGB to YCbCr). The operation of the color conversion unit 40006 according to an embodiment can be optionally applied according to the color value included in the attribute.

[0142] The geometry reconstruction unit 40005 according to the embodiment reconstructs (decompresses) the octree and / or the approximate octree. The geometry reconstruction unit 40005 reconstructs the octree / voxel based on the result of analyzing the distribution of the points. The reconstructed octree / voxel may be referred to as a reconstructed geometry structure (restored geometry structure).

[0143] The attribute transformation unit 40007 according to the embodiment performs attribute transformation to transform the attribute based on the position and / or the reconstructed geometry for which geometric encoding is not performed. As described above, since the attribute depends on the geometry, the attribute transformation unit 40007 can transform the attribute based on the reconstructed geometry information. For example, based on the position value of the point included in the voxel, the attribute transformation unit 40007 can transform the attribute of the point at the position. As described above, when the position of the voxel center is set based on the position of one or more points included in the voxel, the attribute transformation unit 40007 transforms the attributes of the one or more points. When trisoup geometry encoding is performed, the attribute transformation unit 40007 can transform the attribute based on trisoup geometry encoding.

[0144] The attribute transformation unit 40007 can perform attribute transformation by calculating the average value of the attributes or attribute values ​​(e.g., the color or reflectivity of each point) of neighboring points within a specific position / radius from the position (or position value) of the center of each voxel. The attribute transformation unit 40007 can apply a weight according to the distance from the center to each point when calculating the average value. Therefore, each voxel has a position and a calculated attribute (or attribute value).

[0145] The attribute transformation unit 40007 can search for neighbor points within a specific position / radius from the center of each voxel based on a KD tree or a Morton code. The KD tree is a binary search tree and supports the ability to manage point data structures based on position so that the nearest neighbor search (NNS) can be performed quickly. The Morton code is generated by presenting the coordinates (e.g., (x, y, z)) representing the 3D positions of all points as bit values ​​and mixing the bits. For example, when the coordinates representing the position of the point are (5, 9, 1), the bit values ​​of the coordinates are (0101, 1001, 0001). Mixing the bit values ​​according to the bit index in the order of z, y, and x produces 010001000111. This value is represented as a decimal number 1095. That is, the Morton code value of the point with coordinates (5, 9, 1) is 1095. The attribute transformation unit 40007 can sort the points based on the Morton code value and perform NNS by depth-first traversal processing. After the attribute transform operation, when NNS is required in another transform process for attribute encoding, KD tree or Morton code is used.

[0146] As shown in the figure, the transformed attributes are input to the RAHT unit 40008 and / or the LOD generation unit 40009.

[0147] The RAHT unit 40008 according to an embodiment performs RAHT encoding for predicting attribute information based on the reconstructed geometric information. For example, the RAHT unit 40008 can predict attribute information of a higher-level node in the octree based on attribute information associated with a lower-level node in the octree.

[0148] The LOD generation unit 40009 according to the embodiment generates a level of detail (LOD). The LOD according to the embodiment is the detail level of the point cloud content. As the LOD value decreases, it indicates that the detail level of the point cloud content decreases. As the LOD value increases, it indicates that the detail of the point cloud content increases. Points can be classified by LOD.

[0149] The lifting transform unit 40010 according to the embodiment performs lifting transform coding for transforming the attributes of the point cloud based on the weights. As described above, the lifting transform coding may be optionally applied.

[0150] The coefficient quantization unit 40011 according to the embodiment quantizes the attribute after the attribute encoding based on the coefficient.

[0151] The arithmetic encoder 40012 according to the embodiment encodes the quantized attributes based on arithmetic coding.

[0152] Although not shown in this figure, Figure 4 The elements of the point cloud video encoder may be implemented by hardware, software, firmware, or a combination thereof including one or more processors or integrated circuits configured to communicate with one or more memories included in the point cloud content providing device. The one or more processors may perform the above Figure 4 At least one of the operations and / or functions of the elements of the point cloud video encoder. In addition, one or more processors can operate or execute a set of software programs and / or instructions to perform Figure 4 The operation and / or functionality of the elements of the point cloud video encoder. One or more memories according to an embodiment may include a high-speed random access memory, or include a non-volatile memory (e.g., one or more disk storage devices, flash memory devices, or other non-volatile solid-state storage devices).

[0153] Figure 5 An example of a voxel according to an embodiment is shown.

[0154] Figure 5 2 shows voxels located in a 3D space represented by a coordinate system consisting of three axes, namely, an X-axis, a Y-axis, and a Z-axis. Figure 4 As described, the point cloud video encoder (eg, quantization unit 40001) may perform voxelization. A voxel refers to a 3D cubic space generated when a 3D space is divided into units (unit=1.0) based on axes representing the 3D space (eg, X-axis, Y-axis, and Z-axis). Figure 5 An example of a voxel generated by an octree structure is shown, in which the octree consists of two poles (0, 0, 0) and (2 d ,2 d ,2 d ) is recursively subdivided. A voxel consists of at least one point. The spatial coordinates of the voxel can be estimated based on the positional relationship with the voxel group. As mentioned above, the voxel has properties like the pixels of a 2D image / video (such as color or reflectivity). The details of the voxel are similar to those of the reference Figure 4 The details described are the same, so their description is omitted.

[0155] Figure 6 Examples of octrees and occupancy codes are shown according to an embodiment.

[0156] As reference Figures 1 to 4As described, the point cloud content providing system (point cloud video encoder 10002) or the octree analysis unit 40002 of the point cloud video encoder performs octree geometry encoding (or octree encoding) based on the octree structure to efficiently manage the area and / or position of voxels.

[0157] Figure 6 The upper part of FIG. 1 shows an octree structure. The 3D space of the point cloud content according to an embodiment is represented by axes (e.g., X-axis, Y-axis, and Z-axis) of a coordinate system. The octree structure is formed by recursively subdividing the two poles (0, 0, 0) and (2 d ,2 d ,2 d ) is created by defining the cube axis-aligned bounding box. Here, 2 d can be set to the value of the minimum bounding box that constitutes all points around the point cloud content (or point cloud video). Here, d represents the depth of the octree. The value of d is determined in Formula 1. In Formula 1, (x int n ,y int n ,z int n ) represents the position (or position value) of the quantization point.

[0158] [Formula 1]

[0159]

[0160] like Figure 6 As shown in the middle of the upper part, the entire 3D space can be divided into eight spaces according to the partition. Each divided space is represented by a cube with six faces. Figure 6 As shown in the upper right side of , each of the eight spaces is divided again based on the axis of the coordinate system (e.g., X-axis, Y-axis, and Z-axis). Therefore, each space is divided into eight smaller spaces. The divided smaller spaces are also represented by cubes with six faces. This division scheme is applied until the leaf nodes of the octree become voxels.

[0161] Figure 6 The lower part of shows the octree occupancy code. The occupancy code of the octree is generated to indicate whether each of the eight divided spaces generated by dividing one space contains at least one point. Therefore, a single occupancy code is represented by eight child nodes. Each child node represents the occupancy of the divided space, and the child node has a 1-bit value. Therefore, the occupancy code is represented as an 8-bit code. That is, when at least one point is contained in the space corresponding to the child node, the node is assigned a value of 1. When no point is contained in the space corresponding to the child node (the space is empty), the node is assigned a value of 0. Since Figure 6The occupancy code shown in is 00100001, so it indicates that the space corresponding to the third child node and the eighth child node among the eight child nodes each contains at least one point. As shown in the figure, each of the third child node and the eighth child node has 8 child nodes, and the child nodes are represented by 8-bit occupancy codes. The figure shows that the occupancy code of the third child node is 10000111, and the occupancy code of the eighth child node is 01001111. The point cloud video encoder (e.g., arithmetic encoder 40004) according to the embodiment can perform entropy encoding on the occupancy code. In order to improve compression efficiency, the point cloud video encoder can perform intra-frame / inter-frame encoding on the occupancy code. The receiving device (e.g., receiving device 10004 or point cloud video decoder 10006) according to the embodiment reconstructs the octet tree based on the occupancy code.

[0162] The point cloud video encoder (e.g., octree analysis unit 40002) according to the embodiment can perform voxelization and octree encoding to store the positions of the points. However, the points are not always evenly distributed in the 3D space, so there will be specific areas where there are fewer points. Therefore, it is inefficient to perform voxelization on the entire 3D space. For example, when a specific area contains fewer points, it is not necessary to perform voxelization in the specific area.

[0163] Therefore, for the above-mentioned specific area (or nodes other than the leaf nodes of the octree), the point cloud video encoder according to the embodiment can skip voxelization and perform direct encoding to directly encode the positions of the points included in the specific area. The coordinates of the directly encoded points according to the embodiment are called direct coding mode (DCM). The point cloud video encoder according to the embodiment can also perform trisoup geometric encoding based on the surface model to reconstruct the positions of the points in the specific area (or node) based on voxels. Trisoup geometric encoding is a geometric encoding that represents an object as a series of triangular meshes. Therefore, the point cloud video decoder can generate a point cloud from the mesh surface. Trisoup geometric encoding and direct encoding according to the embodiment can be selectively performed. In addition, trisoup geometric encoding and direct encoding according to the embodiment can be performed in combination with octree geometric encoding (or octree encoding).

[0164] In order to perform direct encoding, the option of using direct mode to apply direct encoding should be enabled. The node to which direct encoding will be applied is not a leaf node, and there should be fewer points than a threshold within a specific node. In addition, the total number of points to which direct encoding will be applied should not exceed a preset threshold. When the above conditions are met, the point cloud video encoder (or arithmetic encoder 40004) according to the implementation method can perform entropy encoding on the position (or position value) of the point.

[0165] A point cloud video encoder (e.g., surface approximation analysis unit 40003) according to an embodiment may determine a specific level of the octree (a level less than the depth d of the octree), and may perform trisoup geometry encoding using a surface model starting from that level to reconstruct the position of points in the region of the node based on voxels (trisoup mode). A point cloud video encoder according to an embodiment may specify the level at which trisoup geometry encoding will be applied. For example, when a specific level is equal to the depth of the octree, the point cloud video encoder does not operate in trisoup mode. In other words, a point cloud video encoder according to an embodiment may operate in trisoup mode only when the specified level is less than the depth value of the octree. A 3D cubic area of ​​a node at a specified level according to an embodiment is referred to as a block. A block may include one or more voxels. A block or voxel may correspond to a brick. The geometric structure is represented as a surface within each block. A surface according to an embodiment may intersect at most once with each edge of a block.

[0166] A block has 12 edges, so there are at least 12 intersections in a block. Each intersection is called a vertex (or top point). When there is at least one occupied voxel adjacent to the edge among all blocks sharing the edge, the vertex existing along the edge is detected. An occupied voxel according to an embodiment refers to a voxel containing a point. The position of the vertex detected along the edge is the average position of the edge of all voxels adjacent to the edge among all blocks sharing the edge.

[0167] Once the vertex is detected, the point cloud video encoder according to the embodiment can perform entropy coding on the starting point (x, y, z) of the edge, the direction vector (Δx, Δy, Δz) of the edge, and the vertex position value (relative position value within the edge). When trisoup geometry coding is applied, the point cloud video encoder according to the embodiment (e.g., geometry reconstruction unit 40005) can generate a restored geometry structure (reconstructed geometry structure) by performing triangle reconstruction, upsampling, and voxelization.

[0168] The vertices at the edge of the block determine the surface passing through the block. The surface according to the embodiment is a non-planar polygon. In the triangle reconstruction process, the surface represented by the triangle is reconstructed based on the starting point of the edge, the direction vector of the edge, and the position value of the vertex. According to Formula 2, the triangle reconstruction process is performed by the following operations: ① calculating the centroid value of each vertex, ② subtracting the center value from each vertex value, and ③ estimating the sum of the squares of the values ​​obtained by subtraction.

[0169] [Formula 2]

[0170]

[0171] Then, the minimum value of the sum is estimated, and the projection process is performed according to the axis with the minimum value. For example, when the element x is the smallest, each vertex is projected onto the x-axis relative to the center of the block and onto the (y, z) plane. When the value obtained by projecting onto the (y, z) plane is (ai, bi), the value of θ is estimated by atan2(bi, ai), and the vertices are sorted according to the value of θ. Table 1 below shows the vertex combination for creating a triangle according to the number of vertices. The vertices are sorted from 1 to n. Table 1 below shows that for four vertices, two triangles can be constructed according to the combination of vertices. The first triangle can be composed of vertices 1, 2, and 3 among the sorted vertices, and the second triangle can be composed of vertices 3, 4, and 1 among the sorted vertices.

[0172] Table 1. The triangle formed by the vertices sorted in 1,…,n [Table 1]

[0173]

[0174]

[0175] An upsampling process is performed to add points in the middle along the edges of the triangle and perform voxelization. The added points are generated based on the upsampling factor and the width of the block. The added points are called refinement vertices. The point cloud video encoder according to an embodiment may voxelize the refinement vertices. In addition, the point cloud video encoder may perform attribute encoding based on the voxelized position (or position value). Figure 7 An example of a neighbor node pattern according to an embodiment is illustrated. In order to improve the compression efficiency of the point cloud video, the point cloud video encoder according to the embodiment may perform entropy coding based on context adaptive arithmetic coding.

[0176] As reference Figures 1 to 6 Described, Figure 1 Point cloud content providing system or point cloud video encoder 10002 or Figure 4 The point cloud video encoder or arithmetic encoder 40004 can immediately perform entropy encoding on the occupancy code. In addition, the point cloud content providing system or the point cloud video encoder can perform entropy encoding (intra-frame encoding) based on the occupancy code of the current node and the occupancy of the neighboring nodes, or perform entropy encoding (inter-frame encoding) based on the occupancy code of the previous frame. The frame representation according to the embodiment is a collection of point cloud videos generated simultaneously. The compression efficiency of the intra-frame coding / inter-frame coding according to the embodiment may depend on the number of neighboring nodes referenced. When the number of bits increases, the operation becomes complicated, but the encoding can be biased to one side, which can increase the compression efficiency. For example, when a 3-bit context is given, 2 bits need to be used. 3= 8 ways to perform encoding. The division into parts for encoding affects the complexity of the implementation. Therefore, an appropriate level of compression efficiency and complexity must be met.

[0177] Figure 7 The process of obtaining an occupancy pattern based on the occupancy of neighbor nodes is illustrated. The point cloud video encoder according to the embodiment determines the occupancy of neighbor nodes of each node of the octree and obtains the value of the neighbor pattern. The neighbor node pattern is used to infer the occupancy pattern of the node. Figure 7 The upper part of the diagram shows a cube corresponding to the node (the cube in the middle) and six cubes (neighboring nodes) that share at least one face with the cube. The nodes shown in the diagram are nodes at the same depth. The numbers shown in the diagram represent the weights (1, 2, 4, 8, 16, and 32) associated with the six nodes, respectively. The weights are assigned in sequence according to the positions of the neighboring nodes.

[0178] Figure 7 The lower part shows the neighbor node mode value. The neighbor node mode value is the sum of the values ​​multiplied by the weights of the occupied neighbor nodes (neighbor nodes with points). Therefore, the neighbor node mode value is 0 to 63. When the neighbor node mode value is 0, it indicates that there is no node with a point (unoccupied node) among the neighbor nodes of the node. When the neighbor node mode value is 63, it indicates that all neighbor nodes are occupied nodes. As shown in the figure, since the neighbor nodes assigned weights 1, 2, 4 and 8 are occupied nodes, the neighbor node mode value is 15, which is the sum of 1, 2, 4 and 8. The point cloud video encoder can perform encoding according to the neighbor node mode value (for example, when the neighbor node mode value is 63, 64 encodings can be performed). According to an embodiment, the point cloud video encoder can reduce the encoding complexity by changing the neighbor node mode value (for example, based on a table through which 64 is changed to 10 or 6).

[0179] Figure 8 An example of point configuration in each LOD according to an embodiment is illustrated.

[0180] As reference Figures 1 to 7 As described, before performing attribute encoding, the encoded geometry is reconstructed (decompressed). When direct encoding is applied, the geometry reconstruction operation may include changing the placement of directly encoded points (e.g., placing directly encoded points in front of the point cloud data). When trisoup geometry encoding is applied, the geometry reconstruction process is performed by triangle reconstruction, upsampling, and voxelization. Since the attributes depend on the geometry, attribute encoding is performed based on the reconstructed geometry.

[0181] The point cloud video encoder (e.g., LOD generation unit 40009) can classify (reorganize or group) points by LOD. Figure 8The point cloud content corresponding to the LOD is shown. Figure 8 The leftmost image in the figure shows the original point cloud content. Figure 8 The second picture from the left shows the distribution of points in the lowest LOD, and Figure 8 The rightmost picture in the figure shows the distribution of points in the highest LOD. That is, the points in the lowest LOD are sparsely distributed, and the points in the highest LOD are densely distributed. That is, as the LOD increases from Figure 8 As you go up in the direction indicated by the arrow at the bottom, the space (or distance) between the points becomes narrower.

[0182] Fig. 9 An example of point configuration for each LOD according to an embodiment is illustrated.

[0183] As reference Figures 1 to 8 As described, a point cloud content providing system or a point cloud video encoder (e.g., Figure 1 Point cloud video encoder 10002, Figure 4 The point cloud video encoder or LOD generation unit 40009 can generate LOD. LOD is generated by reorganizing points into a set of refinement levels according to the set LOD distance value (or a set of Euclidean distances). The LOD generation process is performed not only by the point cloud video encoder, but also by the point cloud video decoder.

[0184] Fig. 9 The upper part of shows examples of points (P0 to P9) of point cloud content distributed in 3D space. Fig. 9 In , the original order means the order of points P0 to P9 before LOD generation. Fig. 9 In , the LOD-based order means the order of points generated according to LOD. Points are reorganized by LOD. In addition, a high LOD contains points belonging to a lower LOD. Fig. 9 As shown in , LOD0 contains P0, P5, P4, and P2. LOD1 contains the points of LOD0, P1, P6, and P3. LOD2 contains the points of LOD0, the points of LOD1, P9, P8, and P7.

[0185] As reference Figure 4 As described, the point cloud video encoder according to the embodiment may selectively or in combination perform LOD-based prediction transform coding, LOD-based lifting transform coding, and RAHT transform coding.

[0186] The point cloud video encoder according to the embodiment may generate a predictor for a point to perform LOD-based predictive transform coding to set a prediction attribute (or a prediction attribute value) for each point. That is, N predictors may be generated for N points. The predictor according to the embodiment may calculate a weight (=1 / distance) based on the LOD value of each point, index information about neighboring points that exist within a set distance for each LOD, and the distance to the neighboring point.

[0187] The predicted attribute (or attribute value) according to the embodiment is set to the average value of the value obtained by multiplying the attribute (or attribute value) of the neighboring point set in the predictor of each point (for example, color, reflectivity, etc.) by the weight (or weight value) calculated based on the distance to each neighboring point. The point cloud video encoder (for example, the coefficient quantization unit 40011) according to the embodiment can quantize and inverse quantize the residual (which can be called residual attribute, residual attribute value, attribute prediction residual value or prediction error attribute value, etc.) of each point obtained by subtracting the predicted attribute (or attribute value) of each point from the attribute of each point (that is, the original attribute value). The quantization processing performed on the residual attribute value in the transmitting device is configured as shown in Table 2. The inverse quantization processing performed on the residual attribute value in the receiving device is configured as shown in Table 3.

[0188] [Table 2]

[0189] int PCCQuantization(int value,int quantStep){ if(value>=0){ return floor(value / quantStep+1.0 / 3.0); }else{ return-floor(-value / quantStep+1.0 / 3.0); } }

[0190] [Table 3]

[0191] int PCCInverseQuantization(int value,int quantStep){ if(quantStep==0){ return value; }else{ return value*quantStep; } }

[0192] When the predictor of each point has neighboring points, the point cloud video encoder (e.g., arithmetic encoder 40012) according to the embodiment may perform entropy coding on the residual attribute value after quantization and inverse quantization as described above. When the predictor of each point has no neighboring points, the point cloud video encoder (e.g., arithmetic encoder 40012) according to the embodiment may perform entropy coding on the attributes of the corresponding points without performing the above operations. The point cloud video encoder (e.g., lifting transform unit 40010) according to the embodiment may generate a predictor for each point, set the calculated LOD and register the neighboring points into the predictor, and set weights according to the distance to the neighboring points to perform lifting transform coding. The lifting transform coding according to the embodiment is similar to the above-mentioned predictive transform coding, but the difference is that the weights are cumulatively applied to the attribute values. The process of cumulatively applying weights to the attribute values ​​according to the embodiment is configured as follows.

[0193] 1) Create an array Quantization Weight (QW) for storing the weight value of each point. The initial value of all elements of QW is 1.0. Multiply the QW value of the predictor index of the neighbor node registered in the predictor by the weight of the predictor of the current point, and add the values ​​obtained by the multiplication.

[0194] 2) Boosting prediction processing: A value obtained by multiplying the attribute value of a point by a weight is subtracted from the existing attribute value to calculate a predicted attribute value.

[0195] 3) Create a temporary array called updateweight, and update and initialize it to zero.

[0196] 4) The weight calculated by multiplying the weight calculated for all predictors by the weight corresponding to the predictor index stored in QW is cumulatively added to the update weight array as the index of the neighbor node. The value obtained by multiplying the attribute value of the index of the neighbor node by the calculated weight is cumulatively added to the update array.

[0197] 5) Boosting update process: Divide the attribute value of the update array for all predictors by the weight value of the update weight array indexed by the predictor, and add the existing attribute value to the value obtained by the division.

[0198] 6) The predicted attribute is calculated by multiplying the attribute value updated by the lifting update process by the weight updated by the lifting prediction process (stored in QW) for all predictors. The point cloud video encoder (e.g., coefficient quantization unit 40011) according to the embodiment quantizes the predicted attribute value. In addition, the point cloud video encoder (e.g., arithmetic encoder 40012) performs entropy encoding on the quantized attribute value.

[0199] A point cloud video encoder according to an embodiment (e.g., RAHT unit 40008) can perform RAHT transform coding, in which attributes associated with lower-level nodes in the octree are used to predict attributes of higher-level nodes. RAHT transform coding is an example of attribute intra-frame coding performed by backward scanning of the octree. A point cloud video encoder according to an embodiment scans the entire area from voxels and repeats a merging process of merging voxels into larger blocks in each step until the root node is reached. The merging process according to an embodiment is performed only on occupied nodes. The merging process is not performed on empty nodes. The merging process is performed in a higher mode just above the empty node.

[0200] The following formula 3 represents the RAHT transformation matrix. In formula 3, Represents the average attribute value of the voxels at level l. It can be based on and To calculate and The weight is and

[0201] [Formula 3]

[0202]

[0203] here, is the low-pass value and is used in the merging process at the next higher level. Denotes the high-pass coefficient. The high-pass coefficient in each step is quantized and undergoes entropy coding (eg, encoded by arithmetic encoder 40012). The weight is calculated as As shown in formula 4, and Compute the root node.

[0204] [Formula 4]

[0205]

[0206] The values ​​of gDC are also quantized and undergo entropy coding like the high-pass coefficients.

[0207] Fig.10 A point cloud video decoder according to an embodiment is illustrated.

[0208] Fig.10 The point cloud video decoder illustrated in Figure 1 An example of a point cloud video decoder 10006 described in Figure 1 The point cloud video decoder 10006 illustrated in FIG. 10006 is the same as or similar to the operation of the point cloud video decoder 10006 illustrated in FIG. 10006. As shown in the figure, the point cloud video decoder can receive a geometry bitstream and an attribute bitstream contained in one or more bitstreams. The point cloud video decoder includes a geometry decoder and an attribute decoder. The geometry decoder performs geometry decoding on the geometry bitstream and outputs the decoded geometry structure. The attribute decoder performs attribute decoding on the attribute bitstream based on the decoded geometry and outputs the decoded attributes. The decoded geometry structure and the decoded attributes are used to reconstruct the point cloud content (decoded point cloud).

[0209] Fig.11 A point cloud video decoder according to an embodiment is illustrated.

[0210] Fig.11 The point cloud video decoder illustrated in Fig.10 An example of a point cloud video decoder is illustrated in FIG. 1 and may be executed as Figures 1 to 9 The decoding operation is the inverse of the encoding operation of the point cloud video encoder illustrated in FIG.

[0211] As reference Figure 1 and Fig.10 As described, the point cloud video decoder can perform geometry decoding and attribute decoding. Geometry decoding is performed before attribute decoding.

[0212] According to an embodiment, a point cloud video decoder includes an arithmetic decoder (arithmetic decoding) 11000, an octree synthesizer (synthesized octree) 11001, a surface approximation synthesizer (synthesized surface approximation) 11002 and a geometry reconstruction unit (reconstruction geometry) 11003, an inverse coordinate transformer (inverse transformation coordinates) 11004, an arithmetic decoder (arithmetic decoding) 11005, an inverse quantization unit (inverse quantization) 11006, a RAHT transformer 11007, an LOD generator (generate LOD) 11008, an inverse lifting unit (inverse lifting) 11009 and / or an inverse color transformation unit (inverse transformation color) 11010.

[0213] The arithmetic decoder 11000, the octree synthesizer 11001, the surface approximation synthesizer 11002, the geometry reconstruction unit 11003, and the coordinate inverse transformer 11004 may perform geometry decoding. The geometry decoding according to the embodiment may include direct decoding and trisoup geometry decoding. Direct decoding and trisoup geometry decoding are selectively applied. The geometry decoding is not limited to the above examples and is used as a reference. Figures 1 to 9 This is performed by inverse processing of the geometric encoding described.

[0214] The arithmetic decoder 11000 according to the embodiment decodes the received geometry bitstream based on arithmetic coding. The operation of the arithmetic decoder 11000 corresponds to the inverse process of the arithmetic encoder 40004.

[0215] The octree synthesizer 11001 according to the embodiment can generate an octree by acquiring an occupancy code from a decoded geometry bitstream (or information about a geometry structure protected as a result of decoding). Figures 1 to 9 Configure the occupancy code in detail.

[0216] When trisoup geometry encoding is applied, the surface approximation synthesizer 11002 according to an embodiment may synthesize a surface based on the decoded geometry and / or the generated octree.

[0217] According to an embodiment, the geometry reconstruction unit 11003 may regenerate the geometry based on the surface and / or the decoded geometry. Figures 1 to 9As described, direct coding and trisoup geometry coding are selectively applied. Therefore, the geometry reconstruction unit 11003 directly imports and adds position information about the points to which direct coding is applied. When trisoup geometry coding is applied, the geometry reconstruction unit 11003 can reconstruct the geometry structure by performing the reconstruction operations (e.g., triangle reconstruction, upsampling, and voxelization) of the geometry reconstruction unit 40005. Details and References Figure 6 The details of the description are the same, so the description thereof is omitted. The reconstructed geometry may include a point cloud picture or frame that does not contain attributes.

[0218] The coordinate inverse transformer 11004 according to an embodiment may acquire the position of a point by transforming the coordinates based on the reconstructed geometric structure.

[0219] The arithmetic decoder 11005, the inverse quantization unit 11006, the RAHT transformer 11007, the LOD generation unit 11008, the inverse lifter 11009 and / or the inverse color transform unit 11010 may perform the reference Fig.10 Attribute decoding described. Attribute decoding according to an embodiment includes regional adaptive hierarchical transform (RAHT) decoding, hierarchical nearest neighbor prediction based on interpolation (prediction transform) decoding, and hierarchical nearest neighbor prediction based on interpolation with an update / lifting step (lifting transform) decoding. The above three decoding schemes can be selectively used, or a combination of one or more decoding schemes can be used. Attribute decoding according to an embodiment is not limited to the above examples.

[0220] The arithmetic decoder 11005 according to the embodiment decodes the attribute bit stream through arithmetic coding.

[0221] The inverse quantization unit 11006 according to the embodiment inversely quantizes the information about the decoded attribute bitstream or attribute protected as a decoding result, and outputs the inversely quantized attribute (or attribute value). Inverse quantization may be selectively applied based on attribute encoding of the point cloud video encoder.

[0222] According to an embodiment, the RAHT transformer 11007, the LOD generation unit 11008 and / or the inverse lifter 11009 may process the reconstructed geometry and the inverse quantized attributes. As described above, the RAHT transformer 11007, the LOD generation unit 11008 and / or the inverse lifter 11009 may selectively perform a decoding operation corresponding to the encoding of the point cloud video encoder.

[0223] The inverse color transform unit 11010 according to the embodiment performs inverse transform encoding to inversely transform the color value (or texture) included in the decoded attribute. The operation of the inverse color transform unit 11010 may be selectively performed based on the operation of the color transform unit 40006 of the point cloud video encoder.

[0224] Although not shown in this figure, Fig.11 The elements of the point cloud video decoder may be implemented by hardware, software, firmware, or a combination thereof including one or more processors or integrated circuits configured to communicate with one or more memories included in the point cloud content providing device. The one or more processors may perform the above Fig.11 At least one or more of the operations and / or functions of the elements of the point cloud video decoder. In addition, one or more processors can operate or execute a set of software programs and / or instructions to perform Fig.11 The operation and / or functionality of the elements of the point cloud video decoder.

[0225] Fig.12 A transmitting device according to an embodiment is illustrated.

[0226] Fig.12 The transmitting device shown in Figure 1 The sending device 10000 (or Figure 4 Example of a point cloud video encoder). Fig.12 The sending device illustrated in the example can execute the same as reference Figures 1 to 9 One or more of the operations and methods that are the same or similar to the operations and methods of the point cloud video encoder described. According to an embodiment, a sending device may include a data input unit 12000, a quantization processor 12001, a voxelization processor 12002, an octree occupancy code generator 12003, a surface model processor 12004, an intra-frame / inter-frame coding processor 12005, an arithmetic encoder 12006, a metadata processor 12007, a color transformation processor 12008, an attribute transformation processor 12009, a LOD / lifting / RAHT transformation processor 12010, an arithmetic encoder 12011 and / or a sending processor 12012.

[0227] The data input unit 12000 according to the embodiment receives or acquires point cloud data. The data input unit 12000 may perform the same operation and / or acquisition method as the point cloud video acquisition unit 10001 (or refer to Figure 2 The same or similar operations and / or acquisition methods as described in the acquisition process 20000).

[0228] The data input unit 12000, the quantization processor 12001, the voxelization processor 12002, the octree occupancy code generator 12003, the surface model processor 12004, the intra / inter coding processor 12005 and the arithmetic encoder 12006 perform geometric coding. Figures 1 to 9 The geometric encoding described is the same or similar, so a detailed description thereof is omitted.

[0229] According to an embodiment, the quantization processor 12001 quantizes the geometric structure (e.g., the position value of a point). The operation and / or quantization of the quantization processor 12001 are similar to the reference Figure 4 The operation and / or quantization of the quantization unit 40001 described above are the same or similar. Figures 1 to 9 The details of the description are the same.

[0230] The voxelization processor 12002 according to the embodiment performs voxelization on the quantized position value of the point. The voxelization processor 12002 may perform the same as the reference Figure 4 The operation and / or voxelization process of the quantization unit 40001 described above is the same as or similar to the operation and / or process of the voxelization process described above. Figures 1 to 9 The details of the description are the same.

[0231] The octree occupancy code generator 12003 according to the embodiment performs octree encoding on the voxelized position of the point based on the octree structure. The octree occupancy code generator 12003 can generate an occupancy code. The octree occupancy code generator 12003 can perform the same as the reference Figure 4 and Figure 6 The operations and / or methods of the point cloud video encoder (or octree analysis unit 40002) described above are the same or similar to the operations and / or methods. Figures 1 to 9 The details of the description are the same.

[0232] According to an embodiment, the surface model processor 12004 may perform trisoup geometric encoding based on the surface model to reconstruct the position of a point in a specific area (or node) based on voxels. Figure 4 The operations and / or methods of the point cloud video encoder (e.g., the surface approximation analysis unit 40003) described above are the same or similar to the operations and / or methods. Figures 1 to 9 The details of the description are the same.

[0233] According to the embodiment, the intra-frame / inter-frame encoding processor 12005 can perform intra-frame / inter-frame encoding on the point cloud data. The intra-frame / inter-frame encoding processor 12005 can perform the same as the reference Figure 7 The same or similar encoding as described for intra / inter coding. Details and references Figure 7 The details of the description are the same. According to an embodiment, the intra / inter encoding processor 12005 may be included in the arithmetic encoder 12006.

[0234] According to an embodiment, the arithmetic encoder 12006 performs entropy coding on the octree and / or approximate octree of the point cloud data. For example, the coding scheme includes arithmetic coding. The arithmetic encoder 12006 performs the same or similar operations and / or methods as the operations and / or methods of the arithmetic encoder 40004.

[0235] The metadata processor 12007 according to an embodiment processes metadata (e.g., set values) about the point cloud data and provides it to necessary processing processes such as geometry coding and / or attribute coding. In addition, the metadata processor 12007 according to an embodiment can generate and / or process signaling information related to geometry coding and / or attribute coding. The signaling information according to an embodiment can be encoded separately from the geometry coding and / or attribute coding. The signaling information according to an embodiment can be interleaved.

[0236] The color transformation processor 12008, the attribute transformation processor 12009, the LOD / lifting / RAHT transformation processor 12010, and the arithmetic encoder 12011 perform attribute encoding. Figures 1 to 9 The attribute codes described are the same or similar, so a detailed description thereof is omitted.

[0237] The color transform processor 12008 according to an embodiment performs color transform encoding to transform the color value included in the attribute. The color transform processor 12008 may perform color transform encoding based on the reconstructed geometry. The reconstructed geometry is consistent with the reference Figures 1 to 9 In addition, it performs the same Figure 4 The operations and / or methods of the color conversion unit 40006 described above are the same as or similar to the operations and / or methods described above, and detailed descriptions thereof are omitted.

[0238] The attribute transformation processor 12009 according to an embodiment performs attribute transformation to transform attributes based on the reconstructed geometry and / or the location where geometry encoding is not performed. Figure 4 The operation and / or method of the attribute transformation unit 40007 described in the embodiment are the same as or similar to the operation and / or method of the attribute transformation unit 40007 described in the embodiment. The detailed description thereof is omitted. The LOD / lifting / RAHT transformation processor 12010 according to the embodiment may encode the transformed attribute by any one of RAHT coding, prediction transformation coding and lifting transformation coding or a combination thereof. The LOD / lifting / RAHT transformation processor 12010 performs the same as the reference. Figure 4 The operations of the RAHT unit 40008, the LOD generation unit 40009, and the lifting transform unit 40010 described above are the same as or similar to at least one of the operations. In addition, the prediction transform coding, the lifting transform coding, and the RAHT transform coding are the same as those of the reference Figures 1 to 9 Those described are the same, so a detailed description thereof is omitted.

[0239] The arithmetic encoder 12011 according to the embodiment may encode the encoded attribute based on arithmetic coding. The arithmetic encoder 12011 performs the same or similar operation and / or method as that of the arithmetic encoder 40012.

[0240] The transmitting processor 12012 according to the embodiment may transmit each bitstream containing the encoded geometry and / or the encoded attributes and / or metadata (or metadata information), or transmit a bitstream configured with the encoded geometry and / or the encoded attributes and / or metadata. When the encoded geometry and / or the encoded attributes and / or metadata according to the embodiment are configured into a bitstream, the bitstream may include one or more sub-bitstreams. The bitstream according to the embodiment may include signaling information, which includes a sequence parameter set (SPS) for sequence-level signaling, a geometry parameter set (GPS) for signaling of geometry information encoding, an attribute parameter set (APS) for signaling of attribute information encoding, and a tile parameter set (TPS or tile list) for tile-level signaling and slice data. The slice data may include information about one or more slices. A slice according to the embodiment may include a geometry bitstream Geom0. 0 and one or more attribute bitstreams Attr0 0 and Attr1 0 .

[0241] A slice is a sequence of syntax elements that represents all or part of a coded point cloud frame.

[0242] The TPS according to an embodiment may include information about each tile of one or more tiles (for example, height / size information and coordinate information about a bounding box). The geometry bitstream may include a header and a payload. The header of the geometry bitstream according to an embodiment may include a parameter set identifier (geom_parameter_set_id), a tile identifier (geom_tile_id), and a slice identifier (geom_slice_id) included in the GPS, as well as information about the data contained in the payload. As described above, the metadata processor 12007 according to an embodiment may generate and / or process signaling information and send it to the sending processor 12012. According to an embodiment, an element for performing geometry encoding and an element for performing attribute encoding may share data / information with each other, as indicated by the dotted lines. The sending processor 12012 according to an embodiment may perform operations and / or sending methods that are the same or similar to those of the transmitter 10003. Details and References Figure 1 and Figure 2 The details described are the same, so their description is omitted.

[0243] Fig.13 A receiving device according to an embodiment is illustrated.

[0244] Fig.13 The receiving device illustrated in the example is Figure 1 The receiving device 10004 (or Fig.10 and Fig.11 An example of a point cloud video decoder). Fig.13 The receiving device illustrated in the example may execute the same Figures 1 to 11 One or more of the operations and methods of the point cloud video decoder described are the same or similar to the operations and methods.

[0245] The receiving apparatus according to the embodiment includes a receiver 13000, a receiving processor 13001, an arithmetic decoder 13002, an octtree reconstruction processor 13003 based on an occupancy code, a surface model processor (triangle reconstruction, upsampling, voxelization) 13004, an inverse quantization processor 13005, a metadata parser 13006, an arithmetic decoder 13007, an inverse quantization processor 13008, an LOD / lifting / RAHT inverse transform processor 13009, a color inverse transform processor 13010 and / or a renderer 13011. Each element for decoding according to the embodiment may perform an inverse process of the operation of the corresponding element for encoding according to the embodiment.

[0246] The receiver 13000 according to the embodiment receives the point cloud data. The receiver 13000 may perform the same Figure 1 The operation and / or receiving method of the receiver 10005 is the same as or similar to the operation and / or receiving method of the receiver 10005. A detailed description thereof is omitted.

[0247] The reception processor 13001 according to an embodiment may obtain a geometry bitstream and / or an attribute bitstream from received data. The reception processor 13001 may be included in the receiver 13000 .

[0248] The arithmetic decoder 13002, the octtree reconstruction processor 13003 based on the occupancy code, the surface model processor 13004 and the inverse quantization processor 13005 can perform geometric decoding. Figures 1 to 10 The geometric decoding described is the same or similar, so a detailed description thereof is omitted.

[0249] The arithmetic decoder 13002 according to the embodiment may decode the geometry bitstream based on arithmetic coding. The arithmetic decoder 13002 performs the same or similar operation and / or encoding as that of the arithmetic decoder 11000.

[0250] The octree reconstruction processor 13003 based on the occupancy code according to the embodiment can reconstruct the octree by obtaining the occupancy code from the decoded geometry bitstream (or the information about the geometry structure protected as a result of decoding). The octree reconstruction processor 13003 based on the occupancy code performs the same or similar operations and / or methods as the operations of the octree synthesizer 11001 and / or the octree generation method. When trisoup geometry coding is applied, the surface model processor 13004 according to the embodiment can perform trisoup geometry decoding and related geometry reconstruction (e.g., triangle reconstruction, upsampling, voxelization) based on the surface model method. The surface model processor 13004 performs the same or similar operations as the operations of the surface approximation synthesizer 11002 and / or the geometry reconstruction unit 11003.

[0251] According to an embodiment, the inverse quantization processor 13005 can inverse quantize the decoded geometric structure.

[0252] According to an embodiment, the metadata parser 13006 can parse the metadata contained in the received point cloud data, for example, setting values. The metadata parser 13006 can pass the metadata for geometry decoding and / or attribute decoding. Fig.12 The metadata described is the same, so a detailed description thereof is omitted.

[0253] The arithmetic decoder 13007, the inverse quantization processor 13008, the LOD / RAHT inverse transform processor 13009 and the color inverse transform processor 13010 perform attribute decoding. Figures 1 to 10 The attribute decoding described is the same or similar, so a detailed description thereof is omitted.

[0254] The arithmetic decoder 13007 according to the embodiment can decode the attribute bitstream by arithmetic coding. The arithmetic decoder 13007 can decode the attribute bitstream based on the reconstructed geometric structure. The arithmetic decoder 13007 performs the same or similar operation and / or coding as the operation and / or coding of the arithmetic decoder 11005.

[0255] According to an embodiment, the inverse quantization processor 13008 may inverse quantize the decoded attribute bitstream. The inverse quantization processor 13008 performs the same or similar operations and / or methods as those of the inverse quantization unit 11006 and / or inverse quantization methods.

[0256] The LOD / lifting / RAHT inverse transform processor 13009 according to an embodiment can process the reconstructed geometry and the inverse quantized attributes. The prediction / lifting / RAHT inverse transform processor 1301 performs one or more of the same or similar operations and / or decoding as the RAHT transformer 11007, the LOD generation unit 11008 and / or the inverse lifter 11009. The color inverse transform processor 13010 according to an embodiment performs inverse transform encoding to inverse transform the color value (or texture) included in the decoded attribute. The color inverse transform processor 13010 performs the same or similar operations and / or inverse transform encoding as the inverse color transform unit 11010. The renderer 13011 according to an embodiment can render point cloud data.

[0257] Fig.14 An exemplary structure operably connectable with a method / apparatus for transmitting and receiving point cloud data according to an embodiment is shown.

[0258] Fig.14 The structure of represents a configuration in which at least one of the server 17600, the robot 17100, the autonomous driving vehicle 17200, the XR device 17300, the smart phone 17400, the home appliance 17500, and / or the head mounted display (HMD) 17700 is connected to the cloud network 17000. The robot 17100, the autonomous driving vehicle 17200, the XR device 17300, the smart phone 17400, or the home appliance 17500 is referred to as a device. In addition, the XR device 17300 may correspond to a point cloud compressed data (PCC) device according to an embodiment, or may be operably connected to a PCC device.

[0259] The cloud network 17000 may represent a network that constitutes a part of a cloud computing infrastructure or exists in a cloud computing infrastructure. Here, the cloud network 17000 may be configured using a 3G network, a 4G or Long Term Evolution (LTE) network, or a 5G network.

[0260] The server 17600 may be connected to at least one of the robot 17100 , the autonomous driving vehicle 17200 , the XR device 17300 , the smart phone 17400 , the home appliance 17500 , and / or the HMD 17700 via the cloud network 17000 , and may assist at least a portion of the processing of the connected devices 17100 to 17700 .

[0261] HMD 17700 represents one of the implementation types of an XR device and / or a PCC device according to an embodiment. The HMD type device according to an embodiment includes a communication unit, a control unit, a memory, an I / O unit, a sensor unit, and a power supply unit.

[0262] In the following, various embodiments of apparatuses 17100 to 17500 to which the above technology is applied will be described. According to the above embodiments, Fig.14 the apparatuses 17100 to 17500 illustrated in [the above] can be operably connected / coupled to a point cloud data transmitting device and a receiver.

[0263] <PCC+XR>

[0264] An XR / PCC apparatus can adopt PCC technology and / or XR (AR+VR) technology, and can be implemented as an HMD, a head-up display (HUD) installed in a vehicle, a television, a mobile phone, a smartphone, a computer, a wearable device, a household appliance, a digital signage, a vehicle, a stationary robot, or a mobile robot.

[0265] The XR / PCC apparatus can analyze 3D point cloud data or image data obtained through various sensors or from an external device, and generate position data and attribute data regarding 3D points. Thereby, the XR / PCC apparatus can obtain information about the surrounding space or real objects, and render and output an XR object. For example, the XR / PCC apparatus can match an XR object including auxiliary information about the recognized object with the recognized object, and output the matched XR object.

[0266] <PCC+Autopilot+XR>

[0267] An autonomous vehicle 17200 can be implemented as a mobile robot, a vehicle, an unmanned aerial vehicle, etc. by applying PCC technology and XR technology.

[0268] An autonomous vehicle 17200 applying XR / PCC technology can mean an autonomous vehicle provided with a device for providing an XR image or an autonomous vehicle that is a control / interaction target in an XR image. Specifically, an autonomous vehicle 17200 that is a control / interaction target in an XR image can be distinguished from an XR device 17300, and can be operably connected to the XR device 17300.

[0269] An autonomous vehicle 17200 having a device for providing an XR / PCC image can obtain sensor information from a sensor including a camera, and output a generated XR / PCC image based on the obtained sensor information. For example, the autonomous vehicle 17200 can have an HUD and output an XR / PCC image thereto, thereby providing an XR / PCC object corresponding to a real object or an object existing on a screen to an occupant.

[0270] When the XR / PCC object is output to the HUD, at least a portion of the XR / PCC object may be output to overlap with the real object to which the occupant's eyes are directed. On the other hand, when the XR / PCC object is output to a display disposed inside the autonomous driving vehicle, at least a portion of the XR / PCC object may be output to overlap with the object on the screen. For example, the autonomous driving vehicle 17200 may output XR / PCC objects corresponding to objects such as roads, another vehicle, traffic lights, traffic signs, two-wheeled vehicles, pedestrians, and buildings.

[0271] Virtual reality (VR) technology, augmented reality (AR) technology, mixed reality (MR) technology, and / or point cloud compression (PCC) technology according to an embodiment are applicable to various devices.

[0272] In other words, VR technology is a display technology that only provides CG images of real-world objects, backgrounds, etc. On the other hand, AR technology refers to a technology that shows a virtually created CG image on an image of a real object. MR technology is similar to the above-mentioned AR technology in that the virtual object to be shown is mixed and combined with the real world. However, MR technology is different from AR technology in that AR technology clearly distinguishes between real objects and virtual objects created as CG images, and uses virtual objects as supplementary objects to real objects, while MR technology regards virtual objects as objects with equivalent characteristics to real objects. More specifically, an example of the application of MR technology is a hologram service.

[0273] Recently, VR, AR, and MR technologies are sometimes referred to as extended reality (XR) technologies without being clearly distinguished from each other. Therefore, embodiments of the present disclosure are applicable to any of VR, AR, MR, and XR technologies. Encoding / decoding based on PCC, V-PCC, and G-PCC technologies are applicable to such technologies.

[0274] The PCC method / apparatus according to the embodiment may be applied to a vehicle providing autonomous driving services.

[0275] Vehicles providing autonomous driving services are connected to the PCC device for wired / wireless communication.

[0276] When a point cloud compressed data (PCC) sending / receiving device according to an embodiment is connected to a vehicle for wired / wireless communication, the device can receive / process content data related to AR / VR / PCC services that can be provided together with autonomous driving services and send it to the vehicle. In the case where the PCC sending / receiving device is installed on the vehicle, the PCC sending / receiving device can receive / process content data related to AR / VR / PCC services based on a user input signal input through a user interface device and provide it to the user. The vehicle or user interface device according to the embodiment can receive the user input signal. The user input signal according to the embodiment may include a signal indicating an autonomous driving service.

[0277] In addition, the point cloud video encoder of the sender can also perform a spatial segmentation process of spatially segmenting (or dividing) the point cloud data into one or more 3D blocks before encoding the point cloud data. That is, in order to perform the encoding and sending operations of the sending device and the decoding and rendering operations of the receiving device in real time and with low latency, the sending device can spatially segment the point cloud data into multiple regions. In addition, the sending device can encode the spatially segmented regions (or blocks) independently or non-independently, thereby achieving random access and parallel encoding in the three-dimensional space occupied by the point cloud data. In addition, the sending device and the receiving device can independently or non-independently perform encoding and decoding for each spatially segmented region (or block), thereby preventing the accumulation of errors in the encoding process and the decoding process.

[0278] Fig.15 is a diagram illustrating another example of a point cloud transmitting apparatus according to an embodiment including a space divider.

[0279] The point cloud transmitting device according to the embodiment may include a data input unit 51001, a coordinate transformation unit 51002, a quantization processor 51003, a space divider 51004, a signaling processor 51005, a geometry encoder 51006, an attribute encoder 51007, and a transmitting processor 51008. According to the embodiment, the coordinate transformation unit 51002, the quantization processor 51003, the space divider 51004, the geometry encoder 51006, and the attribute encoder 51007 may be referred to as a point cloud video encoder.

[0280] The data input unit 51001 can execute Figure 1 Some or all operations of the point cloud video acquisition unit 10001, or can be performed Fig.12 The coordinate transformation unit 51002 can perform some or all operations of the data input unit 12000. Figure 4 In addition, the quantization processor 51003 may perform some or all operations of the coordinate transformation unit 40000. Figure 4Some or all of the operations of the quantization unit 40001, or may perform Fig.12 Some or all operations of the quantization processor 12001.

[0281] The spatial splitter 51004 can spatially split the point cloud data output from the quantization processor 51003 after quantization into one or more 3D blocks based on the bounding box and / or sub-bounding box. Here, the 3D block can refer to a tile group, a tile, a slice, a coding unit (CU), a prediction unit (PU) or a transform unit (TU). In one embodiment, the signaling information for spatial segmentation is entropy encoded by the signaling processor 51005 and then sent in the form of a bit stream through the sending processor 51008.

[0282] Fig.16 (a) to Fig.16 (c) of FIG. 1 illustrates an implementation of dividing the bounding box into one or more tiles. Fig.16 As shown in (a), the point cloud object corresponding to the point cloud data can be represented in the form of a box based on a coordinate system, which is called a bounding box. In other words, the bounding box represents a cube that can contain all points in the point cloud.

[0283] Fig.16 (b) and Fig.16 (c) illustrates an example in which Fig.16 An example in which the bounding box of (a) is segmented into tile 1# and tile 2#, and tile 2# is again segmented into slice 1# and slice 2#.

[0284] In one embodiment, the point cloud content may be a person, multiple persons, an object, or multiple objects such as an actor. In a larger scope, it may be a map for autonomous driving or a map for indoor navigation of a robot. In this case, the point cloud content may be a large amount of locally connected data. In this case, the point cloud content cannot be encoded / decoded at one time, so tile segmentation may be performed before compressing the point cloud content. For example, room #101 in a building may be segmented into one tile, and room #102 in the building may be segmented into another tile. In order to support fast encoding / decoding by applying parallelization to the segmented tiles, the tiles may be segmented (or split) into slices again. This operation may be referred to as slice segmentation (or splitting).

[0285] That is, according to an embodiment, a tile may represent a partial area (e.g., a rectangular cube) of a 3D space occupied by point cloud data. According to an embodiment, a tile may include one or more slices. A tile according to an embodiment may be divided into one or more slices, so that a point cloud video encoder may encode point cloud data in parallel.

[0286] A slice may represent a unit of data (or bitstream) that can be independently encoded by a point cloud video encoder according to an embodiment and / or a unit of data (or bitstream) that can be independently decoded by a point cloud video decoder. A slice may be a collection of data in a 3D space occupied by point cloud data or a collection of some data among point cloud data. A slice according to an embodiment may represent a collection or region of points included in a tile according to an embodiment. According to an embodiment, a tile may be divided into one or more slices based on the number of points included in a tile. For example, a tile may be a collection of points divided by the number of points. According to an embodiment, a tile may be divided into one or more slices based on the number of points, and some data may be split or merged in the segmentation process. That is, a slice may be a unit that can be independently encoded within a corresponding tile. In this way, a tile obtained by spatial segmentation is divided into one or more slices for fast and efficient processing.

[0287] The point cloud video encoder according to the embodiment may encode the point cloud data slice by slice or tile by tile, wherein a tile includes one or more slices. In addition, the point cloud video encoder according to the embodiment may perform different quantization and / or transformation on each tile or each slice.

[0288] The positions of one or more 3D blocks (e.g., slices) spatially partitioned by the spatial partitioner 51004 are output to the geometry encoder 51006, and attribute information (or attributes) are output to the attribute encoder 51007. The position may be position information about a point included in a partition unit (a frame, a block, a tile, a tile group, or a slice), and is referred to as geometry information.

[0289] The geometry encoder 51006 constructs and encodes (i.e., compresses) an octree based on the position output from the spatial partitioner 51004 to output a geometry bitstream. The geometry encoder 51006 may reconstruct an octree and / or an approximate octree and output it to the attribute encoder 51007. The reconstructed octree may be referred to as reconstructed geometry (recovered geometry).

[0290] The attribute encoder 51007 encodes (ie, compresses) the attributes output from the space divider 51004 based on the reconstructed geometry output from the geometry encoder 51006, and outputs an attribute bitstream.

[0291] Fig.17 is a detailed block diagram illustrating another example of a geometry encoder 51006 and an attribute encoder 51007 according to an embodiment.

[0292] Fig.17The voxelization processor 53001, octree generator 53002, geometry information predictor 53003 and arithmetic encoder 53004 of the geometry encoder 51006 can perform Figure 4 Some or all operations of the octree analysis unit 40002, the surface approximation analysis unit 40003, the arithmetic encoder 40004 and the geometric reconstruction unit 40005, or some or all operations of the octree analysis unit 40002, the surface approximation analysis unit 40003, the arithmetic encoder 40004 and the geometric reconstruction unit 40005 can be performed. Fig.12 Some or all operations of the voxelization processor 12002, the octree occupancy code generator 12003 and the surface model processor 12004, the intra / inter coding processor 12005 and the arithmetic encoder 12006.

[0293] Fig.17 The attribute encoder 51007 may include a color transformation processor 53005, an attribute transformation processor 53006, an LOD configuration unit 53007, a neighboring point set configuration unit 53008, an attribute information prediction unit 53009, a residual attribute information quantization processor 53010 and an arithmetic encoder 53011.

[0294] In an embodiment, a quantization processor may be further provided between the spatial divider 51004 and the voxelization processor 53001. The quantization processor quantizes the positions of one or more 3D blocks (eg, slices) spatially divided by the spatial divider 51004. In this case, the quantization processor may perform Figure 4 Some or all of the operations of the quantization unit 40001, or performing Fig.12 When a quantization processor is also provided between the spatial divider 51004 and the voxelization processor 53001, some or all operations of the quantization processor 12001 may be omitted or may not be omitted. Fig.15 Quantization processor 51003.

[0295] The voxelization processor 53001 according to an embodiment performs voxelization based on the position of one or more spatially divided 3D blocks (e.g., slices) or their quantized positions. Voxelization refers to a minimized unit for representing position information in 3D space. A point of a point cloud content (or 3D point cloud video) according to an embodiment may be included in one or more voxels. According to an embodiment, a voxel may include one or more points. In an embodiment, when quantization is performed before voxelization, multiple points may belong to one voxel.

[0296] In this specification, when two or more points are included in one voxel, the two or more points are referred to as duplicate points. That is, in the geometry encoding process, overlapping points may be generated by geometry quantization and voxelization.

[0297] The voxelization processor 53001 according to an embodiment may output duplicate points belonging to one voxel to the octree generator 53002 without merging them, or may merge a plurality of points into one point and output the merged point to the octree generator 53002 .

[0298] The octree generator 53002 according to an embodiment generates an octree based on voxels output from the voxelization processor 53001.

[0299] The geometry information predictor 53003 according to the embodiment predicts and compresses the geometry information based on the octree generated by the octree generator 53002, and outputs the predicted and compressed information to the arithmetic encoder 53004. In addition, the geometry information predictor 53003 reconstructs the geometry based on the position changed by compression, and outputs the reconstructed (or decoded) geometry to the LOD configuration unit 53007 of the attribute encoder 51007. The reconstruction of the geometry information may be performed in a device or component separate from the geometry information predictor 53003. In another embodiment, the reconstructed geometry may also be provided to the attribute transformation processor 53006 of the attribute encoder 51007.

[0300] The color transformation processor 53005 of the attribute encoder 51007 corresponds to Figure 4 Color conversion unit 40006 or Fig.12 The color transform processor 12008 of the embodiment. The color transform processor 53005 according to the embodiment performs color transform encoding of transforming the color value (or texture) included in the attribute provided from the data input unit 51001 and / or the spatial divider 51004. For example, the color transform processor 53005 can transform the format of the color information (for example, from RGB to YCbCr). The operation of the color transform processor 53005 according to the embodiment can be selectively applied according to the color value included in the attribute. In another embodiment, the color transform processor 53005 can perform color transform encoding based on the reconstructed geometry. For details of the geometric reconstruction, refer to Figures 1 to 9 Description.

[0301] According to an embodiment, the attribute transformation processor 53006 may perform attribute transformation of transforming attributes based on a position and / or reconstructed geometry for which geometry encoding has not yet been performed.

[0302] The attribute transformation processor 53006 may be referred to as a recoloring unit.

[0303] The operation of the attribute transformation processor 53006 according to the embodiment may be optionally applied depending on whether the duplicate points are merged. According to the embodiment, the merging of duplicate points may be performed by the voxelization processor 53001 or the octree generator 53002 of the geometry encoder 51006.

[0304] In this specification, when points belonging to one voxel are merged into one point in the voxelization processor 53001 or the octree generator 53002, the attribute transformation processor 53006 performs attribute transformation.

[0305] Attribute transformation processor 53006 performs the Figure 4 Attribute transformation unit 40007 or Fig.12 The operations and / or methods of the attribute transformation processor 12009 are the same as or similar to the operations and / or methods.

[0306] According to an embodiment, the geometry information reconstructed by the geometry information predictor 53003 and the attribute information output from the attribute transformation processor 53006 are provided to the LOD configuration unit 53007 for attribute compression.

[0307] According to an embodiment, the attribute information output from the attribute transform processor 53006 may be compressed based on the reconstructed geometric information through one or a combination of two or more of RAHT encoding, LOD-based prediction transform encoding, and lifting transform encoding.

[0308] Hereinafter, as an embodiment, it is assumed that attribute compression is performed by one or a combination of LOD-based predictive transform coding and lifting transform coding. Therefore, the description of RAHT coding will be omitted. For details of RAHT transform coding, refer to Figures 1 to 9 Description.

[0309] The LOD configuration unit 53007 according to an embodiment generates a level of detail (LOD).

[0310] LOD is the degree of detail of the point cloud content. As the LOD value decreases, the detail of the point cloud content decreases. As the LOD value increases, the detail of the point cloud content increases. The points of the reconstructed geometry (i.e., the reconstructed positions) can be classified according to the LOD.

[0311] In an embodiment, in predictive transform coding and lifting transform coding, points may be divided into LODs and grouped.

[0312] This operation may be referred to as an LOD generation process, and groups with different LODs may be referred to as LOD l Here, l represents LOD and is an integer starting from 0. LOD0 is the set consisting of points with the largest distance between them. As l increases, the points belonging to LOD l The distance between the points decreases.

[0313] When the LOD value configuration unit 53007 generates LOD lWhen the neighboring point set configuration unit 53008 according to the implementation mode is set, the neighboring point set configuration unit 53008 can be based on LOD l The set finds X (>0) nearest neighbor (NN) points in a group with the same or lower LOD (i.e., the distance between nodes is large) and registers them as a neighbor set in the predictor. X is the maximum number of points that can be set as neighbor points. X can be entered as a user parameter (also known as an encoder option) or signaled in the signaling information by the signaling processor 51005 (e.g., the lifting_num_pred_nearest_neighbours field signaled in the APS).

[0314] refer to Fig. 9 As an example, find the neighboring points of P3 belonging to LOD1 in LOD0 and LOD1. For example, when the maximum number of points (X) that can be set as neighboring points is 3, the three nearest neighboring nodes of P3 can be P2, P4, and P6. These three nodes are registered as neighboring sets in the predictor of P3. In an embodiment, among the registered nodes, the neighboring node P4 is closest to P3 in terms of distance, then P6 is the next closest node, and P2 is the farthest node among the three nodes. Here, X=3 is merely an embodiment configured to provide an understanding of the present disclosure. The value of X may be different.

[0315] As described above, all points of the point cloud data may have a predictor respectively.

[0316] The attribute information prediction unit 53009 according to the embodiment predicts the attribute value from the neighboring point registered in the predictor, and obtains the residual attribute value of the corresponding point based on the predicted attribute value. The residual attribute value is output to the residual attribute information quantization processor 53010.

[0317] Next, LOD generation and neighboring point search will be described in detail.

[0318] As described above, for attribute compression, the attribute encoder performs a process of generating LOD based on the points of the reconstructed geometry and searching for the nearest neighboring points of the points to be encoded based on the generated LOD. According to an embodiment, even the attribute decoder of the receiving device performs a process of generating LOD and searching for the nearest neighboring points of the points to be decoded based on the generated LOD.

[0319] According to an embodiment, the LOD configuration unit 53007 may configure one or more LODs using one or more LOD generation methods (or LOD configuration methods).

[0320] According to an embodiment, the LOD generation method used in the LOD configuration unit 53007 may be input as a user parameter, or may be signaled in signaling information through the signaling processor 51005. For example, the LOD generation method may be signaled in the APS.

[0321] According to an embodiment, the LOD generation method may be divided into an octree-based LOD generation method, a distance-based LOD generation method, and a sampling-based LOD generation method.

[0322] Fig.18 is a diagram illustrating an example of generating LOD based on an octree according to an embodiment.

[0323] According to an embodiment, when LOD is generated based on an octree, each depth level of the octree may be matched with each LOD, such as Fig.18 That is, the octree-based LOD generation method is a method of generating LOD using the following characteristics: the details of the point cloud data are gradually increased as the depth level in the octree structure increases (i.e., the direction from the root to the leaf). According to an embodiment, the octree-based LOD configuration can be performed from the root node to the leaf node or vice versa from the leaf node to the root node.

[0324] The distance-based LOD generation method according to the embodiment is a method of arranging points according to a Morton code and generating LOD based on the distance between the points.

[0325] The sampling-based LOD generation method according to the embodiment is a method of arranging points according to the Morton code and classifying each k-th point as a lower LOD that can be a candidate for a neighboring point set. That is, during sampling, each k-th point can be selected and classified as belonging to LOD0 to LOD l-1 The remaining points can be registered in the LOD l That is, the selected points become a candidate group that can be selected to be registered in the LOD l The set of neighboring points of a point in the set. k can be different depending on the content of the point cloud. Here, the point can be a point of the captured point cloud or a point of the reconstructed geometry.

[0326] In this disclosure, based on LOD l Collections and LOD0 to LOD l-1 The corresponding lower candidate set will be called the reserved set or LOD reserved set. LOD0 is the set consisting of points with the largest distance between them. As l increases, the number of points belonging to LOD l The distance between the points decreases.

[0327] Depending on the implementation, LOD lThe points in the set are also arranged based on the Morton code order. l-1 The points included in each of the sets are also arranged based on the Morton code order.

[0328] Fig.19 is a diagram illustrating an example of arranging points of a point cloud in a Morton code order according to an embodiment.

[0329] That is, a Morton code for each point is generated based on the x, y, and z position values ​​of each point of the point cloud. When the Morton codes of the points of the point cloud are generated by this process, the points of the point cloud may be arranged in a Morton code order. According to an embodiment, the points of the point cloud may be arranged in an ascending order of the Morton codes. The order of the points arranged in an ascending order of the Morton codes may be referred to as a Morton order.

[0330] The LOD configuration unit 53007 according to an embodiment may generate LOD by performing sampling on points arranged in a Morton order.

[0331] The LOD configuration unit 53007 generates the LOD by applying at least one of an octree-based LOD generation method, a distance-based LOD generation method, or a sampling-based LOD generation method. l The neighboring point set configuration unit 53008 can be based on LOD l The set is used to search for X (>0) nearest neighbor (NN) points in a group with the same or lower LOD (ie, the distance between nodes is large), and the X NN points are registered in the predictor as a neighbor point set.

[0332] In this case, since a lot of time is consumed to search all points in order to configure a neighboring point set, an implementation method is provided for searching neighboring points within a neighboring point search range that includes only some points. The search range refers to the number of points and may be 128, 256, or another value. According to an implementation method, information about the search range may be set in the neighboring point set configuration unit 53008, or may be input as a user parameter. Alternatively, information about the search range may be signaled in signaling information through the signaling processor 51005 (e.g., a lifting_search_range field signaled in the APS).

[0333] Fig. 20 is a diagram illustrating an example of searching for neighboring points in a search range based on LOD according to an embodiment. The arrows illustrated in the figure indicate the Morton order according to an embodiment.

[0334] exist Fig. 20 In an embodiment, the index list includes the LOD set (ie, LOD l ) and keep the list including LOD-basedl At least one lower LOD set of the set (e.g., LOD0 to LOD l-1 ).

[0335] According to an embodiment, the points of the index list are arranged in ascending order based on the size of the Morton code, and the points of the reserved list are also arranged in ascending order based on the size of the Morton code. Therefore, the first point in the points arranged in the Morton order in the index list and the reserved list has the smallest Morton code size.

[0336] According to the embodiment, the neighboring point set configuration unit 53008 may be configured in the LOD0 to LOD1. l-1 The points of the collection and / or LOD l Among the points in the set, a point having a Morton code closest to the Morton code of point Px among the points that are sequentially located before point Px (i.e., points having a Morton code less than or equal to the Morton code of point Px) is searched to generate a point belonging to LOD l The neighboring point set of the point Px (ie, the point to be encoded or the current point) of the set. In this disclosure, the searched point will be referred to as Pi or the center point.

[0337] According to an embodiment, when searching for a center point Pi, the neighboring point set configuration unit 53008 may search for a point Pi having a Morton code closest to the Morton code of point Px among all points located before point Px, or search for a point Pi having a Morton code closest to the Morton code of point Px among points within the search range. In the present disclosure, the search range may be configured by the neighboring point set configuration unit 53008, or may be input as a user parameter. In addition, information about the search range may be signaled in the signaling information by the signaling processor 51005. The present disclosure provides an embodiment of searching for the center point Pi in a reserved list when the number of LODs is multiple and searching for the center point Pi in an index list when the number of LODs is 1. For example, when the number of LODs is 1, the search range may be determined based on the position of the current point in the list arranged by the Morton code.

[0338] According to an embodiment, the neighboring point set configuration unit 53008 compares the point Px with the point located before the searched (or selected) center point Pi (i.e., before Fig. 20 to the left of the center point in ) and after the searched (or selected) center point Pi (i.e., Fig. 20The distance value between the points belonging to the neighboring point search range (to the right of the center point in ). The neighboring point set configuration unit 53008 can register X (for example, 3) NN points as a neighboring point set. In an embodiment, the neighboring point search range is the number of points. The neighboring point search range according to the embodiment may include one or more points located before (i.e., in front of) and / or after (i.e., after) the center point Pi in the Morton order. X is the maximum number of points that can be registered as neighboring points.

[0339] According to an embodiment, information about the neighboring point search range and information about the maximum number X of points that can be registered as neighboring points can be configured by the neighboring point set configuration unit 53008, or can be input as a user parameter, or signaled in the signaling information through the signaling processor 51005 (for example, the lifting_search_range field and the lifting_num_pred_nearest_neighbours field signaled in the APS). According to an embodiment, the actual search range can be a value obtained by multiplying the value of the lifting_search_range field by 2 and then adding the resulting value to the center point (i.e., (the value of the lifting_search_range field × 2) + the center point), a value obtained by adding the center value to the value of the lifting_search_range field (i.e., the value of the lifting_search_range field + the center point), or the value of the lifting_search_range field. The present disclosure provides an embodiment of searching for neighboring points of a point Px in a reserved list when the number of LODs is multiple and searching for neighboring points of a point Px in an index list when the number of LODs is 1.

[0340] For example, if the number of LODs is 2 or more and the information about the search range (e.g., lifting_search_range field) is 128, the actual search range includes the center point Pi in the retention list arranged by the Morton code, the 128 points before the center point Pi, and the 128 points after the center point Pi. As another example, if the number of LODs is 1 and the information about the search range is 128, the actual search range includes the center point Pi in the index list arranged by the Morton code and the 128 points before the center point Pi. As another example, if the number of LODs is 1 and the information about the search range is 128, the actual search range includes the 128 points before the current point Px in the index list arranged by the Morton code.

[0341] According to an embodiment, the neighboring point set configuration unit 53008 may compare the distance values ​​between the points in the actual search range and the point Px, and register the X NN points as the neighboring point set of the point Px.

[0342] On the other hand, attribute characteristics of content may appear different depending on an object from which the content has been captured, a capturing scheme, or a device used to capture the content.

[0343] Fig.21 (a) and Fig.21 (b) is a diagram illustrating an example of point cloud content according to an embodiment.

[0344] Fig.21 (a) is an example of dense point cloud data obtained by capturing an object with a 3D scanner, and the correlation between the attributes of the object may be high. Fig.21 As shown in (a) of FIG. 5 , even when point cloud data consists of one object, the density of points of the point cloud data can affect the similarity of predictions performed based on neighboring points.

[0345] Fig.21 (b) illustrates an example of sparse point cloud data obtained by capturing a wide area with a LiDAR device, and the attribute correlation between points may be low. An example of dense point cloud data may be a still image, and an example of sparse point cloud data may be a drone or an autonomous vehicle. Fig.21 As shown in (b), when the point cloud data consists of multiple objects, especially when a large area is captured with real-time LiDAR, the density of the points of the captured point cloud data is low. Accordingly, the similarity of neighboring points may be low, which can reduce the compression efficiency.

[0346] In this way, the attribute characteristics of the content may appear different depending on the object from which the content was captured, the capture scheme, or the device used to capture the content. If a neighboring point search is performed to configure a neighboring point set when the attribute is encoded without considering different attribute characteristics, the distances between the corresponding points and the registered neighboring points may not all be adjacent. For example, when X (e.g., 3) NN points are registered as a neighboring point set after comparing the distance values ​​between the points belonging to the neighboring point search range before and after the point Px and the center point Pi, at least one of the X registered neighboring points may be far away from the point Px. In other words, at least one of the X registered neighboring points may not be the actual neighboring point of the point Px.

[0347] This phenomenon is more likely to occur in sparse point cloud data than in dense point cloud data. Fig.21As illustrated in (a) of , if there is an object in the content that has color continuity / similarity and is densely captured, and the captured area is small, then there may be content with high attribute correlation between points even if the points are a little far from the neighboring points. Fig.21 As illustrated in (b) of FIG. 1 , if a relatively wide area is sparsely captured by a LiDAR device, content with color correlation may exist only when the distance to the neighboring point is within a certain range. Fig.21 When encoding the attributes of sparse point cloud data as exemplified in (b), some neighboring points registered by neighboring point search may not be actual neighboring points of the point to be encoded.

[0348] When generating LOD, this phenomenon has a high probability of occurring when the points are arranged in Morton order. That is, although the Morton order quickly arranges adjacent points, due to the zigzag scanning characteristics of the Morton code, jump segments (i.e., Fig.19 In this case, some of the neighboring points registered by searching for neighboring points within the search range may not be the actual NN points of the point to be coded.

[0349] Fig. 22 (a) and Fig. 22 (b) is a diagram illustrating an example of an average distance, a minimum distance, and a maximum distance of points belonging to a neighboring point set according to an embodiment.

[0350] Fig. 22 (a) illustrates an example of the 100th frame of the content called FORD, and Fig. 22 (b) illustrates an example of the first frame of content called QNX.

[0351] Fig. 22 (a) and Fig. 22 (b) is an exemplary embodiment to help those skilled in the art understand. Fig. 22 (a) and Fig. 22 (b) may be different frames of the same content or specific frames of different content.

[0352] For the sake of convenience, Fig. 22 The frame illustrated in (a) is referred to as the first frame, and Fig. 22 The frame illustrated in (b) is referred to as the second frame.

[0353] exist Fig. 22In the first frame of (a), the total number of points is 80265, only one NN point is registered among 4 points, 2 NN points are registered among 60 points, and 3 NN points are registered among 80201 points. That is, the maximum number of points that can be registered as NN points is 3, but it can be 1 or 2 depending on the position of the point in the frame.

[0354] exist Fig. 22 In the second frame of (b), the total number of points is 31279, only one NN point is registered in 3496 points, 2 NN points are registered in 219 points, and 3 NN points are registered in 27564 points.

[0355] exist Fig. 22 The first frame of (a) and Fig. 22 In the second frame of (b), NN1 is the NN point, NN2 is the next NN point, and NN3 is the third NN point. Fig. 22 In the first frame of (a), the maximum distance of NN2 is 1407616 and the maximum distance of NN3 is 1892992, while Fig. 22 In the second frame of (b), the maximum distance of NN2 is 116007040 and the maximum distance of NN3 is 310887696.

[0356] That is, it can be understood that there is a considerable difference in distance depending on the content or frame. Therefore, some of the registered neighboring points may not be actual neighboring points.

[0357] In this way, if the distance difference between the point to be encoded and the registered neighboring point becomes larger, the attribute difference between the two points is likely to be large. In addition, when prediction is performed based on a neighboring point set including these neighboring points and a residual attribute value is obtained, the residual attribute value may increase, resulting in an increase in the bitstream size.

[0358] In other words, the characteristics of the content or frame may have an impact on the configuration of the neighboring point set used for attribute prediction, and this may affect the compression efficiency of the attribute.

[0359] Therefore, in the present disclosure, when configuring a neighboring point set in order to encode the attributes of the point cloud content, the neighboring points are selected by considering the attribute correlation between the points of the content, so that meaningless points are not selected as the neighboring point set. Then, since the residual attribute value is reduced and the bitstream size is reduced, the compression efficiency of the attribute is improved. In other words, the present disclosure uses a method of limiting the points that can be selected as the neighboring point set by considering the attribute correlation between the points of the content to improve the compression efficiency of the attribute.

[0360] The present disclosure provides an implementation for applying a maximum distance of neighboring points (also referred to as a maximum NN distance) when configuring a neighboring point set in an attribute encoding process of point cloud content in order to consider attribute correlations between points of the content.

[0361] The present disclosure provides an implementation for configuring a neighbor point set by applying a search range and / or a maximum NN distance when encoding attributes of point cloud content.

[0362] According to an embodiment, the distance between points used to select the nearest neighbor is less than or equal to the maximum neighboring point distance (also called the maximum nearest neighbor distance).

[0363] For example, as a result of comparing the distance values ​​between the point Px and the points belonging to the neighboring point search range before and after the center point Pi, points farther than the maximum NN distance among the X (e.g., 3) NN points are not registered in the neighboring point set of the point Px. In other words, as a result of comparing the distance values ​​between the point Px and the points belonging to the neighboring point search range before and after the center point Pi, only the points within the maximum NN distance among the X (e.g., 3) NN points are registered in the neighboring point set of the point Px. For example, if one of the three points with the closest distance is farther than the maximum NN distance, the farther point can be excluded, and only the remaining two points can be registered in the neighboring point set of the point Px.

[0364] According to the embodiment, the neighboring point set configuration unit 53008 can apply the maximum NN distance according to the attribute characteristics of the content, so as to achieve the best compression efficiency of the attribute regardless of the attribute characteristics of the content. Here, the maximum NN distance can be calculated differently according to the LOD generation method (octree-based, distance-based, or sampling-based LOD generation method). That is, the maximum NN distance can be applied differently to each LOD.

[0365] In the present disclosure, the maximum NN distance is used interchangeably with the maximum distance of neighboring points or the maximum neighboring point distance. In other words, in the present disclosure, the maximum NN distance, the maximum distance of neighboring points and the maximum neighboring point distance have the same meaning.

[0366] In the present disclosure, the maximum NN distance may be obtained by multiplying a basic neighboring point distance (referred to as a basic distance or a reference distance) by NN_range, as illustrated in Equation 5 below.

[0367] [Formula 5]

[0368] Maximum neighbor distance = basic neighbor distance * NN_range

[0369] In Formula 5, NN_range is a range within which neighboring points can be selected, and may be referred to as a maximum range within which neighboring points can be selected, a maximum neighboring point range, a neighboring point range, or a nearest neighbor range.

[0370] The neighboring point set configuration unit 53008 according to the embodiment may automatically or manually set the NN_range according to the characteristics of the content, or set the NN_range by inputting as a user parameter (also referred to as an encoder option). In addition, information related to the NN_range may be signaled in the signaling information through the signaling processor 51005. The signaling information including the information related to the NN_range may be at least one of an SPS, an APS, a tile parameter set, or an attribute slice header.

[0371] When NN_range is automatically calculated, the attribute decoder of the receiving device may also automatically calculate NN_range and apply it to the neighboring point search.

[0372] When information related to NN_range is signaled in the signaling information, the attribute decoder of the receiving device can calculate the maximum neighboring point distance based on the signaling information. In this case, the neighboring points used in attribute prediction in attribute encoding can be reconstructed and applied to attribute decoding.

[0373] According to an embodiment, the neighboring point set configuration unit 53008 can calculate / configure the basic neighboring point distance by combining one or more of an octree-based method, a distance-based method, a sampling-based method, an average difference-based method for a Morton code for each LOD, and an average distance difference-based scheme for each LOD. According to an embodiment, the method of calculating / configuring the basic neighboring point distance and / or the basic neighboring point distance information can be signaled in the signaling information (e.g., at least one of the SPS, APS, TPS, or attribute slice header) by the signaling processor 51005.

[0374] Next, an embodiment of acquiring a basic neighbor point distance and a maximum neighbor point distance when generating LOD based on an octree will be described.

[0375] like Fig.18 As illustrated in FIG. 5 , when the LOD configuration unit 53007 generates an octree-based LOD, each depth level of the octree can be matched with each LOD.

[0376] According to an embodiment, the neighboring point set configuration unit 53008 may obtain the maximum neighboring point distance based on the octree-based LOD.

[0377] According to an embodiment, when the LOD configuration unit 53007 generates an octree-based LOD, spatial scalability can be supported. With spatial scalability, when the source point cloud is dense, a lower resolution point cloud can be accessed, like a thumbnail with lower decoder complexity and / or smaller bandwidth. For example, the spatial scalability function of the geometry can be provided by a process of encoding or decoding only the occupied bits up to a selected depth level by adjusting the depth level of the octree during encoding / decoding of the geometry. In addition, even during encoding / decoding of the attribute, the spatial scalability function of the attribute can be provided by a process of generating an LOD from a selected depth level of the octree and configuring the points on which the encoding / decoding of the attribute will be performed.

[0378] According to an embodiment, for spatial scalability, the LOD configuration unit 53007 may generate an octree-based LOD, and the neighbor point set configuration unit 53008 may perform a neighbor point search based on the generated octree-based LOD.

[0379] When the neighboring point set configuration unit 53008 according to an embodiment searches for neighboring points in a previous LOD based on points belonging to the current LOD generated based on the octree, the maximum neighboring point distance can be obtained by multiplying the basic neighboring point distance (referred to as the basic distance or reference distance) by NN_range (i.e., basic neighboring point distance × NN_range).

[0380] According to an embodiment, a maximum distance (ie, a diagonal distance) of a node in a specific octree-generated LOD may be defined as a basic distance (or a basic neighboring point distance).

[0381] According to an embodiment, when a LOD in the index list l Set) to search for the previous LOD (for example, LOD0 to LOD in the retention list l-1 ), the diagonal distance of the higher node (parent node) of the octree node of the current LOD can be the basic distance for obtaining the maximum neighboring point distance.

[0382] According to an embodiment, a basic distance which is a diagonal distance (ie, a maximum distance) of one node at a specific LOD may be obtained as in Equation 6 below.

[0383] [Formula 6]

[0384]

[0385] According to another embodiment, a basic distance that is a diagonal distance (ie, a maximum distance) of one node at a specific LOD may be obtained based on L2 as shown in Equation 7 below.

[0386] [Formula 7]

[0387]

[0388] Typically, when calculating the distance between two points, the Manhattan distance calculation method or the Euclidean distance calculation method is used. Manhattan distance will be referred to as L1 distance and Euclidean distance will be referred to as L2 distance. Euclidean space can be defined using Euclidean distance, and the norm corresponding to the distance will be referred to as the Euclidean norm or L2 norm. The norm is a method (function) for measuring the length or size of a vector.

[0389] Fig.23 (a) of FIG. 1 illustrates an example of obtaining the diagonal distance of the octree node of LOD0. According to an embodiment, when Fig.23 When the example of (a) is applied to Equation 6, the basic distance becomes And when Fig.23 When the example of (a) is applied to Formula 7, the basic distance becomes 3.

[0390] Fig.23 (b) illustrates an example of calculating the diagonal distance of the octree node of LOD1. According to an embodiment, when Fig.23 When the example of (b) is applied to Equation 6, the basic distance becomes And when Fig.23 When example (b) is applied to Equation 7, the basic distance becomes 12.

[0391] Fig.24 (a) and Fig.24 (b) illustrates an example of a maximum range (NN_range) that can be selected as a neighboring point at each LOD. More specifically, Fig.24 (a) illustrates an example when NN_range is 1 at LOD0, and Fig.24 (b) illustrates an example when NN_range is 3 at LOD0. According to an embodiment, NN_range may be set automatically or manually according to the characteristics of the content, or may be set by inputting as a user parameter. Information related to NN_range may be signaled in signaling information. The signaling information including information related to NN_range may be at least one of an SPS, an APS, a tile parameter set, or an attribute slice header. According to an embodiment, the range NN_range that may be set to a neighboring point may be set to an arbitrary value (e.g., a multiple of a basic distance) regardless of the octree node range.

[0392] According to an embodiment, when Fig.24 When the example (a) is applied to equations 5 and 6, the maximum neighboring point distance becomes And when Fig.24 When the example (b) is applied to equations 5 and 6, the maximum neighboring point distance becomes

[0393] According to an embodiment, when Fig.24 When the example of (a) is applied to equations 5 and 7, the maximum neighboring point distance becomes 3 (=3*1), and when Fig.24 When the example of (b) is applied to equations 5 and 7, the maximum neighboring point distance becomes 9 (=3*3).

[0394] Next, an embodiment of acquiring a basic neighboring point distance and a maximum neighboring point distance when generating a distance-based LOD will be described.

[0395] According to an embodiment, when configuring a distance-based LOD, a basic neighbor distance may be set to be used to configure the current LOD (ie, LOD l That is, the basic distance of each LOD can be set to the distance (dist2 L ).

[0396] Fig.25 An example of a basic neighboring point distance belonging to each LOD according to an embodiment is illustrated.

[0397] Therefore, when configuring distance-based LOD, you can create a distance-based LOD by placing the distance-based base neighbor at distance dist2 L Multiply by NN_range to get the maximum neighbor distance.

[0398] According to an embodiment, NN_range can be automatically or manually set by the neighboring point set configuration unit 53008 according to the characteristics of the content, or can be set by input as a user parameter. Information about NN_range can be signaled in signaling information. The signaling information including information about NN_range can be at least one of SPS, APS, tile parameter set, or attribute slice header. According to an embodiment, the NN_range that can be set to a neighboring point can be set to an arbitrary value (for example, a multiple of the basic distance). According to an embodiment, the NN_Range can be adjusted according to a method for calculating the distance to the neighboring point.

[0399] Next, an embodiment of acquiring a basic neighbor distance and a maximum neighbor distance when generating a sampling-based LOD will be described.

[0400] According to an embodiment, when generating a sampling-based (i.e., decimation-based) LOD, the basic distance may be determined according to the sampling rate. According to an embodiment, in the sampling-based LOD configuration method, a Morton code is generated based on the position value of the point, the points are arranged according to the Morton code, and then the points that do not correspond to the kth point according to the arrangement order are registered in the current LOD. Therefore, each LOD may have a different k and may be represented as k L .

[0401] According to an embodiment, when LOD is generated by applying a sampling method to points arranged based on Morton codes, the kth consecutive points at each LOD are L The average distance of the points can be set as the basic neighboring point distance at each LOD. That is, the kth consecutive points at the current LOD arranged based on the Morton code L The average distance of the points may be set as the basic neighboring point distance of the current LOD. According to another embodiment, when LOD is generated by applying a sampling method to points arranged based on Morton codes, the depth level of the octree may be estimated from the current LOD level, and the diagonal length of the node of the estimated octree depth level may be set as the basic neighboring point distance. That is, the basic neighboring point distance may be obtained by applying the above formula 6 or formula 7.

[0402] Therefore, when configuring sample-based LOD, the maximum neighbor distance can be obtained by multiplying the sample-based basic neighbor distance by NN_range.

[0403] According to an embodiment, NN_range can be automatically or manually set by the neighboring point set configuration unit 53008 according to the characteristics of the content, or can be set by input as a user parameter. Information about NN_range can be signaled in signaling information. The signaling information including information about NN_range can be at least one of SPS, APS, tile parameter set, or attribute slice header. According to an embodiment, the NN_range that can be set to a neighboring point can be set to an arbitrary value (for example, a multiple of the basic distance). According to an embodiment, the NN_Range can be adjusted according to a method for calculating the distance to the neighboring point.

[0404] In the present disclosure, a method other than the above-mentioned method of acquiring the basic neighboring point distance may be used to obtain the basic neighboring point distance.

[0405] For example, when configuring LOD, the basic neighbor point distance can be configured by calculating the average difference of the Morton code of the sampling points. In addition, the maximum neighbor point distance can be obtained by multiplying the configured basic neighbor point distance by NN_range.

[0406] As another example, whether using a distance-based method or a sampling-based method, the basic neighbor distance can be configured by calculating the average distance difference of the currently configured LOD. In addition, the maximum neighbor distance can be obtained by multiplying the configured basic neighbor distance by NN_range.

[0407] As another example, the basic neighbor distance of each LOD may be received as a user parameter and then applied when obtaining the maximum neighbor distance. Thereafter, information about the received basic neighbor distance may be signaled in the signaling information.

[0408] As described above, the basic neighboring point distance can be obtained by applying the best basic neighboring point distance calculation method according to the LOD configuration method. In another embodiment, the method for calculating the basic neighboring point distance can be applied by selecting any desired method and signaling the selected method in the signaling information so as to restore the point in the decoder of the receiving device using the method. According to the embodiment, a method for calculating the maximum neighboring point distance that is suitable for the properties of the point cloud content and can maximize the compression performance can be used.

[0409] Next, an embodiment of automatically calculating the maximum neighboring point range (NN_range) will be described.

[0410] According to an embodiment, a method of obtaining a maximum neighboring point range may vary according to an LOD generation method.

[0411] According to an embodiment, the neighboring point set configuration unit 53008 may automatically calculate the maximum neighboring point range according to the density, or automatically calculate the maximum neighboring point range by inferring the density according to the length of the axis and the number of points.

[0412] Fig.26 is a flowchart illustrating automatic calculation of a maximum neighboring point range based on density according to an embodiment of the present disclosure.

[0413] First, scan some of the points to calculate the basic neighboring point distance of each LOD (operation 55001). Here, according to the implementation, it is assumed that the basic neighboring point distance of each LOD is greater than 0 (>0). In the present disclosure, for simplicity, the basic neighboring point distance of each LOD is referred to as dist20.

[0414] According to an embodiment, the basic neighboring point distance may be obtained differently according to an LOD generation method.

[0415] For example, when LOD is generated based on an octree, the diagonal distance (i.e., the maximum distance) of a node in a specific LOD may be set as the basic neighboring point distance (e.g., see Formula 6 or Formula 7). According to an embodiment, when there are multiple LODs, the diagonal distance of the upper node (parent node) of the octree node of the current LOD may be set as the basic neighboring point distance.

[0416] For example, when generating LOD based on distance, the basic neighbor distance can be set to be used to generate the LOD based on the current LOD (ie, LOD l That is, the basic neighboring point distance of each LOD can be set to the distance dist2 generated by the LOD of the L level applied to the LOD L (See Fig.25 ).

[0417] For example, when LODs are generated by applying a sampling method to points arranged based on Morton codes, the kth consecutive points in each LOD L The average distance of consecutive points can be set as the basic neighboring point distance at each LOD. That is, the average distance of consecutive k points in the current LOD arranged based on the Morton code L The average distance of the point can be set as the basic neighboring point distance of the current LOD. As another example, when LOD is generated by applying a sampling method to points arranged based on Morton codes, the depth level of the octree can be estimated from the current LOD level, and the diagonal length of the node of the estimated octree depth level can be set as the basic neighboring point distance. In this case, the basic neighboring point distance can be obtained by applying Formula 6 or Formula 7.

[0418] According to an embodiment, the number of points selected for each LOD in order to calculate the basic neighboring point distance may be one quarter, or may be selected in consideration of other intervals.

[0419] Once the basic neighboring point distance is calculated in operation 55001, the diagonal length of the bounding box of the point is calculated (operation 55002). In this disclosure, for simplicity, the diagonal length of the bounding box will be referred to as BBox_diagonal_dist or BBox_diagonal_length.

[0420] According to an embodiment, the bounding box may be set by scanning the point cloud data. According to another embodiment, the bounding box may be inferred and calculated with reference to the octree depth for the octree construction.

[0421] A density rate (or density_estimation) may be estimated based on the basic neighboring point distance calculated in operation 55001 and the diagonal length of the bounding box calculated in operation 55002 (operation 55003).

[0422] In an embodiment, the density rate may be obtained by dividing the basic neighboring point distance by the diagonal length of the bounding box (density_rate=dist20 / BBox_diagonal_dist). In this case, as the density rate increases, the density decreases.

[0423] In another embodiment, the density rate may be obtained by dividing the diagonal length of the bounding box by the basic neighboring point distance (density_rate=BBox_diagonal_dist / dist20). In this case, when the maximum neighboring point range is set according to the density rate, the Kn value and sign may change. In this case, as the density rate decreases, the density decreases. According to an embodiment, the Kn value may be a table value for setting the maximum neighboring point range according to a calculated value obtained by automatically calculating the maximum neighboring point range.

[0424] Once the density rate is obtained in operation 55003, the maximum neighboring point range is set according to the density rate (operation 55004).

[0425] According to an embodiment, when the density rate is greater than Kn (where n>=0 and Kn<1), the value of n or n+1 may be set as the maximum neighboring point range (n: density_rate>Kn (n>=0, Kn<1)).

[0426] According to an embodiment, the Kn value may be a table value for setting the maximum neighboring point range according to the density rate calculated when the maximum neighboring point range is automatically calculated.

[0427] According to an embodiment, the Kn value may be set as a default value according to the LOD generation method, or may be set differently according to user input. That is, information related to the Kn value may be variably changed and signaled in the signaling information, or a fixed value may be preset in the attribute encoder / decoder by default.

[0428] For example, assuming that K1=0.000001 and the density ratio is greater than 0.000001, the maximum neighboring point range (NN_range) is 1.

[0429] According to an embodiment, the maximum neighboring point range information automatically calculated according to the density rate may be signaled in the signaling information. For example, the signaling information including the maximum neighboring point range information may be at least one of an SPS, an APS, a tile parameter set, or an attribute slice header.

[0430] Fig. 27 FIG. 1 is a flowchart illustrating another embodiment of the present disclosure for automatically calculating the maximum neighboring point range based on density. Fig. 27 In an implementation of the invention, the maximum neighbor range is calculated by inferring the density based on the axis length and the number of points.

[0431] According to an embodiment, the density may be inferred based on at least one of a minimum axis length of the bounding box, a maximum axis length of the bounding box, and a median axis length of the bounding box.

[0432] Fig. 27 An implementation of obtaining the maximum neighbor point range based on the minimum axis length of the bounding box and the number of points is illustrated.

[0433] That is, the minimum axis length of the bounding box of the point is calculated (operation 57001). In the present disclosure, for simplicity, the minimum axis length of the bounding box is referred to as BBox_min_axis_length.

[0434] A density rate (density_rate or density_estimation) may be estimated based on the number of points and the minimum axis length of the bounding box calculated in operation 57001 (operation 57002).

[0435] In an embodiment, the density rate may be obtained by dividing the minimum axis length of the bounding box by the number of points (density_rate=BBox_min_axis_length / number of points). In this case, as the density rate increases, the density decreases.

[0436] Once the density rate is obtained in operation 57002, the maximum neighboring point range is set according to the density rate (operation 57003).

[0437] According to an embodiment, when the density rate is greater than Kn (where n>=0 and Kn<1), the value of n or n+1 may be set as the maximum neighboring point range (n: density_rate>Kn (n>=0, Kn<1)).

[0438] According to an embodiment, the Kn value may be a table value for setting the maximum neighboring point range according to the density rate calculated when the maximum neighboring point range is automatically calculated.

[0439] According to an embodiment, the Kn value may be set as a default value according to the LOD generation method, or may be set differently according to user input. That is, information related to the Kn value may be variably changed and signaled in the signaling information, or a fixed value may be preset in the attribute encoder / decoder by default.

[0440] For example, assuming K1=0.05 and the density ratio is greater than 0.05, the maximum neighboring point range (NN_range) is 1.

[0441] According to another embodiment, the density rate can be estimated based on the maximum axis length of the bounding box and the number of points, or can be estimated based on the middle axis length of the bounding box and the number of points. Then, the maximum neighboring point range is calculated based on the density rate. In this case, the method of calculating the maximum neighboring point range is the same as the method of calculating the maximum neighboring point range based on the minimum axis length of the bounding box described above.

[0442] According to an embodiment, a method of calculating a maximum neighboring point range that can match the attributes of the point cloud content and maximize the compression performance may be used.

[0443] According to an embodiment, any desired method for calculating the maximum neighbor range may be selected and applied, or a combination of one or more methods may be applied. Alternatively, information about whether to perform automatic calculation may be signaled and sent in signaling information. According to an embodiment, when the information about whether to perform automatic calculation indicates automatic calculation, the attribute decoder of the receiving device may automatically calculate the maximum neighbor range.

[0444] According to an embodiment, information about a basic neighbor distance calculation method (e.g., nn_base_distance_calculation_method_type), basic neighbor distance information (e.g., nn_base_distance), maximum neighbor range information (e.g., nearest_neighbore_max_range), minimum neighbor range information (e.g., nearest_neighbor_min_range), information about how to apply the maximum neighbor distance (e.g., nn_range_filtering_location_type), information about whether to automatically calculate the maximum neighbor range (e.g., automatic_nn_range_calculation_flag), information about a method for calculating the maximum neighbor range (e.g., automatic_nn_range_method_type), and information related to a Kn value of an automatic maximum neighbor range (e.g., automatic_max_nn_range_in_table, automatic_nn_range_table_k) may be referred to as neighbor selection related option information. The neighbor selection related option information may be signaled and transmitted in signaling information. In an embodiment, the signaling information may be at least one of a sequence parameter set, an attribute parameter set, a tile parameter set, or an attribute slice header.

[0445] According to an embodiment, the neighboring point set configuration unit 53008 and / or the signaling processor 51005 may set the basic neighboring point distance calculation method (e.g., nn_base_distance_calculation_method_type) in the signaling information according to the LOD generation method (or LOD configuration method). According to an embodiment, the basic neighboring point distance calculation method may include an octree-based basic neighboring point distance calculation, a distance-based basic neighboring point distance calculation, a sampling-based basic neighboring point distance calculation, a basic neighboring point distance calculation based on the average difference of Morton codes between LODs, a basic neighboring point distance calculation based on the average distance difference between LODs, and a basic neighboring point distance calculation based on an input basic neighboring point distance.

[0446] According to an embodiment, the basic neighboring point distance calculated according to the basic neighboring point distance calculation method can be applied to the calculation of the maximum neighboring point range. In addition, the basic neighboring point distance calculation method can be signaled and sent in the signaling information. In this case, the attribute decoder can calculate the maximum neighboring point range based on the signaling information and apply it to the maximum neighboring point distance.

[0447] According to an embodiment, the neighboring point set configuration unit 53008 and / or the signaling processor 51005 may receive basic neighboring point distance information (base_distance) as a user parameter (or encoder option) and apply it to the calculation of the maximum neighboring point distance and / or the maximum neighboring point range. In addition, when the basic neighboring point distance information is signaled in the signaling information, the attribute decoder of the receiving device may calculate the maximum neighboring point distance and / or the maximum neighboring point range based on the signaling information.

[0448] According to an embodiment, the neighboring point set configuration unit 53008 and / or the signaling processor 51005 may signal the maximum neighboring point range information in the signaling information, and the attribute decoder of the receiving device may calculate the maximum neighboring point distance based on the signaling information. According to an embodiment, the maximum neighboring point range may be set differently for each LOD.

[0449] According to an embodiment, the neighboring point set configuration unit 53008 and / or the signaling processor 51005 can signal the neighboring point minimum range information in the signaling information, and the attribute decoder of the receiving device can calculate the maximum neighboring point distance based on the signaling information. According to an embodiment, the minimum neighboring point range can be set differently for each LOD.

[0450] According to an embodiment, the neighboring point set configuration unit 53008 and / or the signaling processor 51005 may receive information indicating whether to automatically calculate the maximum neighboring point range or the minimum neighboring point range or both the maximum neighboring point range and the minimum neighboring point range as a user parameter (or referred to as an encoder option), and signal the information indicating whether to perform automatic calculation in the signaling information.

[0451] According to an embodiment, when the neighboring point set configuration unit 53008 and / or the signaling processor 51005 automatically calculates the maximum neighboring point range, it can receive a Kn value as a user parameter (or encoder option) and signal information related to the Kn value in the signaling information. According to an embodiment, the Kn value can be a table value for setting the maximum neighboring point range according to a calculated value obtained by automatically calculating the maximum neighboring point range. According to an embodiment, when the value of the information about whether to perform automatic calculation is TRUE (true), the attribute decoder of the receiving device can use the information related to the Kn value signaled in the signaling information to determine the maximum neighboring point range.

[0452] According to an embodiment, when automatically calculating the maximum neighboring point range, the neighboring point set configuration unit 53008 and / or the signaling processor 51005 may receive information about the method for calculating the maximum neighboring point range as a user parameter (or encoder option). In addition, information about the maximum neighboring point range calculation method (e.g., automatic_nn_range_method_type) may be signaled in the signaling information. According to an embodiment, the method for automatically calculating the maximum neighboring point range may include a method for calculating the maximum neighboring point range based on density (based on the denominator bounding box diagonal length or the numerator of the bounding box diagonal length) and a method for inferring density based on axis length and number of points and calculating the maximum neighboring point range (based on the maximum, minimum or middle axis of the bounding box). According to an embodiment, the neighboring point set configuration unit 53008 may use a method for calculating the maximum neighboring point range based on density and / or a method for inferring density based on axis length and number of points and calculating the maximum neighboring point range as an automatic calculation method to automatically calculate the maximum neighboring point range. According to an embodiment, when the value of the information on whether to perform automatic calculation is TRUE, the attribute decoder of the receiving device may determine the maximum neighboring point range using information on the maximum neighboring point range calculation method signaled in the signaling information.

[0453] According to an embodiment, the neighboring point set configuration unit 53008 may calculate the maximum neighboring point distance based on the base neighboring point distance (base_distance) and the maximum neighboring point range.

[0454] According to an embodiment, the neighboring point set configuring unit 53008 may register only one or more points having a distance shorter than the calculated maximum neighboring point distance as a neighboring point set of a corresponding point.

[0455] According to an embodiment, the neighboring point set configuration unit 53008 may receive a method of applying a maximum neighboring point distance (e.g., nn_range_filtering_location_type) as a user parameter (or encoder option) and signal the method of applying the maximum neighboring point distance in signaling information. According to an embodiment, the method of applying the maximum neighboring point distance may include a method of calculating the distance between points when selecting neighboring points without selecting points outside the maximum neighboring point distance and a method of registering all neighboring points by calculating the distance between points and deleting points exceeding the maximum neighboring point distance from among the registered points. According to an embodiment, the attribute decoder of the receiving device may configure the neighboring point set by information about a method of applying the maximum neighboring point distance signaled in the signaling information (e.g., nn_range_filtering_location_type) to the neighboring point search.

[0456] The above-mentioned LOD configuration method, basic neighboring point distance acquisition method, maximum neighboring point range (NN_range) acquisition method and maximum neighboring point distance acquisition method can be applied in the same or similar manner as when the attribute decoder of the receiving device configures the neighboring point set and LOD to predict the transformation or enhance the transformation.

[0457] According to an embodiment, if the LOD1 set is generated based on the basic neighboring point distance and the maximum neighboring point range (NN_range) as described above and the maximum neighboring point distance is determined, the neighboring point set configuration unit 53008 searches for X (e.g., 3) NN points among the points within the search range in the group having the same or lower LOD (i.e., the distance between the nodes is large) based on the LOD1 set. Then, the neighboring point set configuration unit 53008 can register only the NN points within the maximum neighboring point distance among the X (e.g., 3) NN points as a neighboring point set. Therefore, the number of NN points registered as a neighboring point set is equal to or less than X. In other words, the NN points among the X NN points that are not within the maximum neighboring point distance are not registered as a neighboring point set and are excluded from the neighboring point set. For example, if two of the three NN points are not within the maximum neighboring point distance, the two NN points are excluded and only the other NN point (i.e., one NN point) is registered as a neighboring point set.

[0458] According to another embodiment, if the LOD1 set is generated based on the basic neighbor point distance and the maximum neighbor point range (NN_range) and the maximum neighbor point distance is determined as described above, the neighbor point set configuration unit 53008 can search for X (for example, 3) NN points within the maximum neighbor point distance among the points within the search range in a group with the same or lower LOD (i.e., the distance between nodes is large) based on the LOD1 set, and register the X NN points as a neighbor point set.

[0459] That is, the number of NN points that can be registered as a neighboring point set can vary depending on the timing (or location) of applying the maximum neighboring point distance. In other words, the number of NN points registered as a neighboring point set can vary depending on whether the maximum neighboring point distance is applied after searching for X NN points through a neighboring point search or when the distance between points is calculated. X is the maximum number of neighboring points that can be set and can be input as a user parameter or signaled in the signaling information by the signaling processor 51005 (e.g., the lifting_num_pred_nearest_neighbours field signaled in the APS).

[0460] Fig.28is a diagram illustrating an example of searching for neighboring points by applying a search range and a maximum neighboring point distance according to an embodiment. The arrows illustrated in the figure indicate a Morton order according to an embodiment.

[0461] exist Fig.28 In an embodiment, the index list includes the LOD set (ie, LOD l ) and keep the list including LOD-based l At least one lower LOD set of the set (e.g., LOD0 to LOD l-1 ).

[0462] According to an embodiment, the points in the index list and the points in the reserved list are arranged in ascending order based on the size of the Morton code. Therefore, the first point of the points in the index list and the reserved list arranged in Morton order has the smallest Morton code size.

[0463] According to the embodiment, the neighboring point set configuration unit 53008 may be configured in the LOD0 to LOD1. l-1 Among the points of the set and / or the points belonging to the LOD1 set, a center point Pi having a Morton code closest to the Morton code of the point Px is searched among the points located sequentially before the point Px (i.e., points having a Morton code less than or equal to the Morton code of the point Px) so as to register the points belonging to the LOD l The set of neighboring points of the point Px (i.e., the point to be encoded or the current point).

[0464] According to an embodiment, when searching for a center point Pi, the neighboring point set configuration unit 53008 may search for a point Pi having a Morton code closest to the Morton code of point Px among all points located before point Px, or search for a point Pi having a Morton code closest to point Px among points within the search range. In the present disclosure, the search range may be configured by the neighboring point set configuration unit 53008, or may be input as a user parameter (also referred to as an encoder option). In addition, information related to the search range may be signaled in a signaling message by a signaling processor 51005. The present disclosure provides an embodiment of searching for a center point Pi in a reserved list when the number of LODs is multiple and searching for a center point Pi in an index list when the number of LODs is 1. For example, when the number of LODs is 1, the search range may be determined based on the position of the current point in the list arranged with the Morton code.

[0465] According to the embodiment, the neighboring point set configuration unit 53008 performs a point Px and a point located before the searched (or selected) center point Pi (i.e., before Fig.28 to the left of the center point in ) and after the searched (or selected) center point Pi (i.e., Fig.28The neighboring point search range is compared with the distance values ​​between the points belonging to the neighboring point search range (to the right of the center point in). The neighboring point set configuration unit 53008 can select X (for example, 3) NN points, and only register the points within the maximum neighboring point distance at the LOD to which Px belongs among the selected X points as the neighboring point set of point Px. In an embodiment, the neighboring point search range is the number of points. The neighboring point search range according to the embodiment may include one or more points located before (i.e., in front of) and / or after (i.e., behind) the center point Pi in the Morton order. X is the maximum number of points that can be registered as neighboring points.

[0466] According to an embodiment, information related to the neighboring point search range and information about the maximum number X of points that can be registered as neighboring points can be configured by the neighboring point set configuration unit 53008, or can be input as a user parameter, or signaled in the signaling information through the signaling processor 51005 (for example, the lifting_search_range field and the lifting_num_pred_nearest_neighbours field signaled in the APS). According to an embodiment, the actual search range can be a value obtained by multiplying the value of the lifting_search_range field by 2 and then adding the center point to the obtained value (i.e., (the value of the lifting_search_range field × 2) + the center point), a value obtained by adding the center value to the value of the lifting_search_range field (i.e., the value of the lifting_search_range field + the center point), or the value of the lifting_search_range field. The present disclosure provides an embodiment of searching for neighboring points of a point Px in a reserved list when the number of LODs is multiple and searching for neighboring points of a point Px in an index list when the number of LODs is 1.

[0467] For example, if the number of LODs is 2 or more and the information related to the search range (e.g., lifting_search_range field) is 128, the actual search range includes the center point Pi in the retention list arranged by the Morton code, the 128 points before the center point Pi, and the 128 points after the center point Pi. As another example, if the number of LODs is 1 and the information related to the search range is 128, the actual search range includes the center point Pi in the index list arranged by the Morton code and the 128 points before the center point Pi. As another example, if the number of LODs is 1 and the information related to the search range is 128, the actual search range includes the 128 points before the current point Px in the index list arranged by the Morton code.

[0468] According to an embodiment, the neighboring point set configuration unit 53008 may compare the distance values ​​between the points in the actual search range and the point Px to search for X NN points, and register only the points within the maximum neighboring point distance at the LOD to which the point Px belongs among the X points as the neighboring point set of the point Px. That is, the neighboring points registered as the neighboring point set of the point Px are limited to the points within the maximum neighboring point distance at the LOD to which the point Px belongs among the X points.

[0469] As described above, the present disclosure can provide an effect of enhancing attribute compression efficiency by limiting points that can be selected as a neighboring point set in consideration of attribute characteristics (correlations) between points in terms of point cloud contents.

[0470] Fig.29 2 is a diagram illustrating another example of searching for neighboring points by applying a search range and a maximum neighboring point distance according to an embodiment. The arrows illustrated in the figure indicate a Morton order according to an embodiment.

[0471] because Fig.29 The example is the same as that of Fig.28 The examples are similar, so the repeated description thereof will be omitted in this article. Fig.29 For the omitted or undescribed parts, refer to Fig.28 Description.

[0472] According to an embodiment, the points of the index list and the points of the reserved list are arranged in ascending order based on the size of the Morton code.

[0473] According to the embodiment, the neighboring point set configuration unit 53008 may be configured in the LOD0 to LOD1. l-1 The points of the collection and / or LOD l Among the points in the set, the center point Pi having a Morton code closest to the Morton code of the point Px among the points located in order before the point Px (i.e., the points having a Morton code less than or equal to the Morton code of the point Px) is searched to register the points belonging to the LOD l The set of neighboring points of the point Px (i.e., the point to be encoded or the current point).

[0474] According to the embodiment, the neighboring point set configuration unit 53008 performs a point Px and a point located before the searched (or selected) center point Pi (i.e., before Fig.29 to the left of the center point in ) and after the searched (or selected) center point Pi (i.e., Fig.29 The neighboring point set configuration unit 53008 may select X (e.g., 3) NN points within the maximum neighboring point distance at the LOD to which point Px belongs and register the X points as the neighboring point set of point Px.

[0475] The present disclosure provides an implementation of searching for neighboring points of point Px in an actual search range of a reserved list when the number of LODs is multiple, and searching for neighboring points of point Px in an actual search range of an index list when the number of LODs is 1.

[0476] For example, if the number of LODs is 2 or more and the information about the search range (e.g., lifting_search_range field) is 128, the actual search range includes the center point Pi in the retention list arranged by the Morton code, the 128 points before the center point Pi, and the 128 points after the center point Pi. As another example, if the number of LODs is 1 and the information about the search range is 128, the actual search range includes the center point Pi in the index list arranged by the Morton code and the 128 points before the center point Pi. As another example, if the number of LODs is 1 and the information about the search range is 128, the actual search range includes the 128 points before the current point Px in the index list arranged by the Morton code.

[0477] When comparing the distance values ​​between the points in the actual search range and the point Px, the neighboring point set configuration unit 53008 according to the embodiment selects X (for example, 3) NN points within the maximum neighboring point distance at the LOD to which the point Px belongs from among the points in the actual search range, and registers these X points as the neighboring point set of the point Px. That is, the neighboring points registered as the neighboring point set of the point Px are limited to the points within the maximum neighboring point distance at the LOD to which the point Px belongs among the X points.

[0478] As described above, the present disclosure can provide an effect of enhancing compression efficiency of attributes by selecting a neighboring point set in consideration of attribute characteristics (correlation) between points of point cloud content.

[0479] According to the above-mentioned embodiment, when a neighboring point set is registered in each predictor of a point to be encoded in the neighboring point set configuration unit 53008, the attribute information prediction unit 53009 predicts the attribute value of the corresponding point from one or more neighboring points registered in the predictor. As described above, when configuring the neighboring point set, by applying the maximum neighboring point distance, the number of neighboring points included in the neighboring point set registered in each predictor is equal to or less than X (for example, 3). According to an embodiment, the predictor of a point can register a 1 / 2 distance (=weight) based on the distance value to each neighboring point with the help of the registered neighboring point set. For example, taking (P2 P4 P6) as the neighboring point set, the predictor of the P3 node calculates the weight based on the distance value to each neighboring point. According to an embodiment, the weight of the neighboring point can be

[0480] According to an embodiment, when a neighboring point set is registered in the predictor, the neighboring point set configuration unit 53008 or the attribute information prediction unit 53009 may normalize the weight of each neighboring point by the total weight of the neighboring points included in the neighboring point set.

[0481] For example, the weights of all neighboring points in the neighboring set of node P3 are summed (total_weight = And the weight of each neighboring point is divided by the sum of the weights Thus, the weight of each neighboring point is normalized.

[0482] Then, the attribute information prediction unit 5301 may predict the attribute value through the predictor.

[0483] According to an embodiment, the average of the values ​​obtained by multiplying the attributes (e.g., color, reflectivity, etc.) of the neighboring points registered in the predictor by the weight (or normalized weight) may be set as the prediction result (i.e., the predicted attribute value). Alternatively, the attribute of a specific point may be set as the prediction result (i.e., the predicted attribute value). According to an embodiment, the predicted attribute value may be referred to as predicted attribute information. In addition, the predicted attribute value (or predicted attribute information) of the point may be subtracted from the attribute value (i.e., the original attribute value) of the point to obtain a residual attribute value (or residual attribute information or residual).

[0484] According to an embodiment, a compression result value may be pre-calculated by applying various prediction modes (or predictor indexes), and then a prediction mode (ie, predictor index) generating a minimum bitstream may be selected from among the prediction modes.

[0485] Next, the process of selecting a prediction mode will be described in detail.

[0486] In this specification, the prediction mode has the same meaning as the predictor index (Preindex) and may be broadly referred to as a prediction method.

[0487] In an embodiment, the process of finding the most appropriate prediction mode for each point and setting the found prediction mode in the predictor of the corresponding point may be performed by the attribute information prediction unit 53009.

[0488] According to an embodiment, a prediction mode in which a predicted attribute value is calculated by weighted averaging (i.e., an average obtained by multiplying the attribute of the neighboring points set in the predictor of each point by a weight calculated based on the distance to each neighboring point) will be referred to as prediction mode 0. In addition, a prediction mode in which the attribute of the first neighboring point is set to the predicted attribute value will be referred to as prediction mode 1, a prediction mode in which the attribute of the second neighboring point is set to the predicted attribute value will be referred to as prediction mode 2, and a prediction mode in which the attribute of the third neighboring point is set to the predicted attribute value will be referred to as prediction mode 3. In other words, a value of the prediction mode (or predictor index) equal to 0 may indicate that the attribute value is predicted by weighted averaging, and a value equal to 1 may indicate that the attribute value is predicted by the first neighboring node (i.e., neighboring point). A value equal to 2 may indicate that the attribute value is predicted by the second neighboring node, and a value equal to 3 may indicate that the attribute value is predicted by the third neighboring node.

[0489] According to an embodiment, the residual attribute value in prediction mode 0, the residual attribute value in prediction mode 1, the residual attribute value in prediction mode 2, and the residual attribute value in prediction mode 3 may be calculated, and a score or a double score may be calculated based on each residual attribute value. Then, the prediction mode with the lowest calculated score may be selected and set as the prediction mode of the corresponding point.

[0490] According to an embodiment, when a preset condition is met, a process of searching for the most suitable prediction mode among a plurality of prediction modes and setting it as the prediction mode of the corresponding point may be performed. Therefore, when the preset condition is not met, a fixed (or predefined) prediction mode (e.g., prediction mode 0) in which the prediction attribute value is calculated by weighted average may be set as the prediction mode of the point without performing the process of searching for the most suitable prediction mode. In an embodiment, this process is performed for each point.

[0491] According to an embodiment, when the difference of the attribute elements (e.g., R, G, B) between the neighboring points registered in the predictor of the point is greater than or equal to a preset threshold value (e.g., lifting_adaptive_prediction_threshold), or when the difference of the attribute elements (e.g., R, G, B) between the neighboring points registered in the predictor of the point is calculated and the sum of the maximum difference of the elements is greater than or equal to the preset threshold value, the preset condition may be satisfied for a specific point. For example, it is assumed that point P3 is a specific point, and points P2, P4, and P6 are registered as neighboring points of point P3. In addition, it is assumed that when the difference of the R, G, and B values ​​between points P2 and P4, the difference of the R, G, and B values ​​between points P2 and P6, and the difference of the R, G, and B values ​​between points P4 and P6 are calculated, the maximum difference of the R value is obtained between points P2 and P4, the maximum difference of the G value is obtained between points P4 and P6, and the maximum difference of the B value is obtained between points P2 and P6. In addition, it is assumed that among the maximum difference in R value (between P2 and P4), the maximum difference in G value (between P4 and P6), and the maximum difference in B value (between P2 and P6), the difference in R value between points P2 and P4 is the largest.

[0492] Under these assumptions, when the R value difference between points P2 and P4 is greater than or equal to a preset threshold, or when the sum of the R value difference between points P2 and P4, the G value difference between points P4 and P6, and the B value difference between points P2 and P6 is greater than or equal to the preset threshold, a process of searching for the most appropriate prediction mode among a plurality of candidate prediction modes may be performed. In addition, a prediction mode (e.g., predIndex) may be signaled only when the R value difference between points P2 and P4 is greater than or equal to a preset threshold, or when the sum of the R value difference between points P2 and P4, the G value difference between points P4 and P6, and the B value difference between points P2 and P6 is greater than or equal to a preset threshold.

[0493] According to another embodiment, when the maximum difference between the values ​​of the attributes (e.g., reflectivity) of the neighboring points registered in the predictor of a specific point is greater than or equal to a preset threshold (e.g., lifting_adaptive_prediction_threshold), the preset condition may be satisfied for the point. In addition, it is assumed that among the reflectivity difference between points P2 and P4, the reflectivity difference between points P2 and P6, and the reflectivity difference between points P4 and P6, the reflectivity difference between points P2 and P4 is the largest.

[0494] Under this assumption, when the reflectivity difference between points P2 and P4 is greater than or equal to a preset threshold, a process of finding the most suitable prediction mode among multiple candidate prediction modes can be performed. In addition, a prediction mode (e.g., predIndex) can be signaled only when the reflectivity difference between points P2 and P4 is greater than or equal to a preset threshold.

[0495] According to an embodiment, the selected prediction mode (e.g., predIndex) of the corresponding point may be signaled in the attribute slice data. In this case, the sender sends the residual attribute value obtained based on the selected prediction mode, and the receiver obtains the predicted attribute value of the corresponding point based on the signaled prediction mode, and adds the predicted attribute value and the received residual attribute value to restore the attribute value of the corresponding point.

[0496] In another embodiment, when the prediction mode is not signaled, the sender may calculate the predicted attribute value based on the prediction mode set as the default mode (e.g., prediction mode 0), and calculate and send the residual attribute value based on the difference between the original attribute value and the predicted attribute value. The receiver may calculate the predicted attribute value based on the prediction mode set as the default mode (e.g., prediction mode 0), and restore the attribute value by adding the value to the received residual attribute value.

[0497] According to an embodiment, the threshold value may be signaled in the signaling information or directly input by the signaling processor 51005 (eg, the lifting_adaptive_prediction_threshold field signaled in the APS).

[0498] According to an embodiment, when a preset condition is satisfied for a specific point as described above, a predictor candidate may be generated. The predictor candidate is referred to as a prediction mode or a predictor index.

[0499] According to an embodiment, prediction modes 1 to 3 may be included in the predictor candidates. According to an embodiment, prediction mode 0 may be included or may not be included in the predictor candidates. According to an embodiment, at least one prediction mode not mentioned above may also be included in the predictor candidates.

[0500] The prediction mode set for each point through the above-mentioned processing and the residual attribute value in the set prediction mode are output to the residual attribute information quantization processor 53010.

[0501] According to an embodiment, the residual property information quantization processor 53010 may apply zero run length encoding to the input residual property value.

[0502] According to an embodiment of the present disclosure, quantization and zero-run-length encoding may be performed on the residual property value.

[0503] According to an embodiment, the arithmetic encoder 53011 applies arithmetic coding to the residual property value and the prediction mode output from the residual property information quantization processor 53010, and outputs the result as a property bitstream.

[0504] The geometry bit stream compressed and output by the geometry encoder 51006 and the attribute bit stream compressed and output by the attribute encoder 51007 are output to the sending processor 51008.

[0505] According to the embodiment, the transmission processor 51008 may execute Fig.12 The operations and / or sending methods of the sending processor 12012 are the same or similar to the operations and / or sending methods, and perform the same Figure 1 The operation and / or transmission method of the transmitter 10003 is the same as or similar to the operation and / or transmission method. Figure 1 or Fig.12 Description.

[0506] According to an embodiment, the transmission processor 51008 can separately transmit the geometry bit stream output from the geometry encoder 51006, the attribute bit stream output from the attribute encoder 51007, and the signaling bit stream output from the signaling processor 51005, or can multiplex the bit streams into one bit stream to be transmitted.

[0507] The transmission processor 51008 according to an embodiment may encapsulate the bitstream in a file or a segment (eg, a streaming segment), and then transmit the encapsulated bitstream through various networks such as a broadcast network and / or a broadband network.

[0508] The signaling processor 51005 according to the embodiment may generate and / or process signaling information and output it to the transmission processor 51008 in the form of a bit stream. The signaling information generated and / or processed by the signaling processor 51005 will be provided to the geometry encoder 51006, the attribute encoder 51007 and / or the transmission processor 51008 for geometry encoding, attribute encoding and transmission processing. Alternatively, the signaling processor 51005 may receive the signaling information generated by the geometry encoder 51006, the attribute encoder 51007 and / or the transmission processor 51008.

[0509] In the present disclosure, signaling information may be signaled and sent in units of parameter sets (sequence parameter set (SPS), geometry parameter set (GPS), attribute parameter set (APS), tile parameter set (TPS), etc.). In addition, it may be signaled and sent based on the coding unit of each image such as a slice or tile. In the present disclosure, the signaling information may include metadata (e.g., setting values) related to the point cloud data, and may be provided to the geometry encoder 51006, the attribute encoder 51007, and / or the sending processor 51008 for geometry encoding, attribute encoding, and sending processing. Depending on the application, the signaling information may also be defined on the system side such as a file format, dynamic adaptive streaming over HTTP (DASH), or MPEG Media Transport (MMT), or on the wired interface side such as a High-Definition Multimedia Interface (HDMI), a display port, a Video Electronics Standards Association (VESA), or a CTA.

[0510] The method / apparatus according to the embodiment may signal relevant information to add / perform the operation of the embodiment. The signaling information according to the embodiment may be used in the sending device and / or the receiving device.

[0511] According to an embodiment of the present disclosure, information about the maximum number of predictors to be used for attribute prediction (lifting_max_num_direct_predictors), threshold information (lifting_adaptive_prediction_threshold) for enabling adaptive prediction of attributes, neighbor selection related option information, etc. can be signaled in at least one of a sequence parameter set, an attribute parameter set, a tile parameter set, or an attribute slice header. In addition, according to an embodiment, predictor index information (predIndex) indicating a prediction mode corresponding to a predictor candidate selected from a plurality of predictor candidates can be signaled in attribute slice data.

[0512] According to an embodiment, the neighbor selection related option information may include information about a basic neighbor distance calculation method (e.g., nn_base_distance_calculation_method_type), basic neighbor distance information (e.g., nn_base_distance), maximum neighbor range information (e.g., nearest_neighbore_max_range), minimum neighbor range information (e.g., nearest_neighbor_min_range), information about a method for applying the maximum neighbor distance (e.g., nn_range_filtering_location_type), information about whether to automatically calculate the maximum neighbor range (e.g., automatic_nn_range_calculation_flag), information about a method for calculating the maximum neighbor range (e.g., automatic_nn_range_method_type), and at least one of information related to the Kn value (e.g., automatic_max_nn_range_in_table, automatic_nn_range_table_k). According to an embodiment, the neighboring point selection related option information may further include at least one of information about the maximum number of points that can be set as neighboring points (e.g., lifting_num_pred_nearest_neighbours), search range related information (e.g., lifting_search_range), and / or LOD configuration methods.

[0513] Similar to the point cloud video encoder of the sending device, the point cloud video decoder of the receiving device performs and generates LOD l Collection, LOD-based l The decoder may further include a method for performing the same or similar processing as the method for performing the normalization of the point by using the prediction mode received by the decoder. The method may further include a ...

[0514] Fig.30 is a diagram illustrating another exemplary point cloud receiving apparatus according to an embodiment.

[0515] The point cloud receiving device according to the embodiment may include a receiving processor 61001, a signaling processor 61002, a geometry decoder 61003, an attribute decoder 61004, and a post-processor 61005. According to the embodiment, the geometry decoder 61003 and the attribute decoder 61004 may be referred to as a point cloud video decoder. According to the embodiment, the point cloud video decoder may be referred to as a PCC decoder, a PCC decoding unit, a point cloud decoder, a point cloud decoding unit, etc.

[0516] The receiving processor 61001 according to the embodiment may receive a single bitstream, or may receive a geometry bitstream, an attribute bitstream, and a signaling bitstream separately. When receiving a file and / or a segment, the receiving processor 61001 according to the embodiment may decapsulate the received file and / or segment and output the decapsulated file and / or segment as a bitstream.

[0517] When a single bitstream is received (or decapsulated), the receiving processor 61001 according to the embodiment can demultiplex the geometry bitstream, the attribute bitstream and / or the signaling bitstream from the single bitstream. The receiving processor 61001 can output the demultiplexed signaling bitstream to the signaling processor 61002, output the geometry bitstream to the geometry decoder 61003, and output the attribute bitstream to the attribute decoder 61004.

[0518] When the geometry bitstream, attribute bitstream and / or signaling bitstream are received (or decapsulated) respectively, the receiving processor 61001 according to the implementation manner can transmit the signaling bitstream to the signaling processor 61002, the geometry bitstream to the geometry decoder 61003, and the attribute bitstream to the attribute decoder 61004.

[0519] The signaling processor 61002 may parse signaling information from the input signaling bit stream, for example, information contained in SPS, GPS, APS, TPS, metadata, etc., process the parsed information, and provide the processed information to the geometry decoder 61003, the attribute decoder 61004, and the post-processor 61005. In another embodiment, the signaling information contained in the geometry slice header and / or the attribute slice header may also be parsed by the signaling processor 61002 before the corresponding slice data is decoded. That is, when Fig.16 As shown, when the sending side divides the point cloud data into tiles and / or slices, the TPS includes the number of slices included in each tile, and accordingly, the point cloud video decoder according to the embodiment can check the number of slices and quickly parse the information for parallel decoding.

[0520] Therefore, the point cloud video decoder according to the present disclosure can quickly parse the bitstream containing point cloud data when it receives the SPS with reduced data volume. The receiving device can decode the tile after receiving the tile, and can decode each slice based on the GPS and APS included in each tile. Thus, the decoding efficiency can be maximized.

[0521] That is, the geometry decoder 61003 can perform a decode operation on the compressed geometry bitstream based on the signaling information (e.g., geometry-related parameters). Fig.15 The geometry is reconstructed by performing the inverse process of the operation of the geometry encoder 51006. The geometry restored (or reconstructed) by the geometry decoder 61003 is provided to the attribute decoder 61004. The attribute decoder 61004 can perform the following operations on the compressed attribute bitstream based on the signaling information (e.g., attribute-related parameters) and the reconstructed geometry: Fig.15 The attributes are restored by reversing the operation of the attribute encoder 51007. Fig.16 As shown in , when the sending side divides the point cloud data into tiles and / or slices, the geometry decoder 61003 and the attribute decoder 61004 perform geometry decoding and attribute decoding on a tile-by-tile and / or slice-by-slice basis.

[0522] Fig.31 is a detailed block diagram illustrating another example of a geometry decoder 61003 and an attribute decoder 61004 according to an embodiment.

[0523] Fig.31 The arithmetic decoder 63001, the octree reconstruction unit 63002, the geometry information prediction unit 63003, the inverse quantization processor 63004 and the coordinate inverse transformation unit 63005 included in the geometry decoder 61003 can perform Fig.11 Some or all of the operations of the arithmetic decoder 11000, the octree synthesis unit (octree synthesizer) 11001, the surface approximation synthesis unit 11002, the geometry reconstruction unit 11003, and the coordinate inverse transformation unit (coordinate inverse transformer) 11004 in the embodiment of the present invention may be performed, or some or all of the operations of the arithmetic decoder 11000, the octree synthesis unit (octree synthesizer) 11001, the surface approximation synthesis unit 11002, the geometry reconstruction unit 11003, and the coordinate inverse transformation unit (coordinate inverse transformer) 11004 may be performed. Fig.13 The position restored by the geometry decoder 61003 is output to the post-processor 61005.

[0524] According to an embodiment, when information about the maximum number of predictors to be used for attribute prediction (lifting_max_num_direct_predictors), threshold information for enabling adaptive prediction of the attribute (lifting_adaptive_prediction_threshold), neighboring point selection related information, etc. are signaled in at least one of a sequence parameter set (SPS), an attribute parameter set (APS), a tile parameter set (TPS), or an attribute slice header, they can be acquired by the signaling processor 61002 and provided to the attribute decoder 61004, or can be directly acquired by the attribute decoder 61004.

[0525] According to an embodiment, the attribute decoder 61004 may include an arithmetic decoder 63006, an LOD configuration unit 63007, a neighboring point set configuration unit 63008, an attribute information prediction unit (attribute information predictor) 63009, a residual attribute information inverse quantization processor 63010 and an inverse color transform processor 63011.

[0526] The arithmetic decoder 63006 according to the embodiment may perform arithmetic decoding on the input attribute bitstream. The arithmetic decoder 63006 may decode the attribute bitstream based on the reconstructed geometry. The arithmetic decoder 63006 performs the same Fig.11 The arithmetic decoder 11005 or Fig.13 The operation and / or decoding of the arithmetic decoder 13007 are the same or similar operations and / or decoding.

[0527] According to an embodiment, the attribute information output from the arithmetic decoder 63006 may be decoded based on the reconstructed geometric information through one or a combination of two or more of RAHT decoding, LOD-based prediction transform decoding, and lifting transform decoding.

[0528] This is described as an embodiment in which the transmitting device performs attribute compression by one or a combination of LOD-based predictive transform encoding and lifting transform encoding. Therefore, an embodiment in which the receiving device performs attribute decoding by one or a combination of LOD-based predictive transform decoding and lifting transform decoding will be described. The description of RAHT decoding of the receiving device will be omitted.

[0529] According to an embodiment, the attribute bitstream arithmetically decoded by the arithmetic decoder 63006 is provided to the LOD configuration unit 63007. According to an embodiment, the attribute bitstream provided from the arithmetic decoder 63006 to the LOD configuration unit 63007 may contain a prediction mode and a residual attribute value.

[0530] The LOD configuration unit 63007 according to the embodiment generates one or more LODs in the same or similar manner as the LOD configuration unit 53007 of the transmitting device, and outputs the generated one or more LODs to the neighboring point set configuration unit 63008.

[0531] According to an embodiment, the LOD configuration unit 63007 may configure one or more LODs using one or more LOD generation methods (or LOD configuration methods). According to an embodiment, the LOD generation method used in the LOD configuration unit 63007 may be provided by the signaling processor 61002. For example, the LOD generation method may be signaled in the APS of the signaling information. According to an embodiment, the LOD generation method may be divided into an octree-based LOD generation method, a distance-based LOD generation method, and a sampling-based LOD generation method.

[0532] According to an embodiment, groups with different LODs are referred to as LOD l Here, l represents LOD, which is an integer starting from 0. LOD0 is the set consisting of points with the largest distance between them. As l increases, the number of points belonging to LOD l The distance between the points decreases.

[0533] According to an embodiment, the prediction mode and the residual property value encoded by the transmitting device may be provided for each LOD or only for a leaf node.

[0534] In one embodiment, when the LOD configuration unit 63007 generates LOD l When collecting, the neighboring point collection configuration unit 63008 can be based on LOD l The set searches for neighboring points equal to or less than X (eg, 3) points in a group having the same or lower LOD (ie, the distance between nodes is large), and registers the searched neighboring points in the predictor as a neighboring point set.

[0535] According to an embodiment, the neighbor point set configuration unit 63008 configures the neighbor point set by obtaining a search range and / or a maximum neighbor point distance based on signaling information and / or applying the search range and / or the maximum neighbor point distance to the neighbor point search process.

[0536] According to the embodiment, the neighboring point set configuration unit 63008 can obtain the maximum neighboring point distance by multiplying the basic neighboring point distance by NN_range. NN_range is a range within which neighboring points can be selected, and will be referred to as the maximum range within which neighboring points can be selected, the maximum neighboring point range, the neighboring point range, or the nearest neighbor range. According to the embodiment, the neighboring point set configuration unit 63008 can calculate / configure the basic neighboring point distance by combining one or more of an octree-based method, a distance-based method, a sampling-based method, a Morton code average difference-based method for each LOD, and an average distance difference-based method for each LOD based on signaling information.

[0537] Since a detailed description of calculating / configuring the basic neighboring point distance by combining one or more of an octree-based method, a distance-based method, a sampling-based method, a Morton code average difference-based method for each LOD, and an average distance difference-based method for each LOD in the neighboring point set configuration unit 63008 according to an embodiment has been given in detail in the process of encoding attributes in the above-mentioned sending device, its details will be omitted in this article.

[0538] According to an embodiment, the neighboring point set configuration unit 63008 may acquire the basic neighboring point distance from the neighboring point selection related option information included in the signaling information.

[0539] According to an embodiment, the neighboring point set configuration unit 63008 may automatically or manually set NN_range according to the characteristics of the content. The neighboring point set configuration unit 63008 may automatically or manually set NN_range according to the characteristics of the content based on the neighboring point selection related option information included in the signaling information.

[0540] According to an embodiment, the neighboring point set configuration unit 63008 may directly obtain the maximum neighboring point range (NN_range) from the neighboring point selection related option information, or may calculate the NN_range based on the neighboring point selection related option information.

[0541] The details of automatically calculating NN_range by the neighboring point set configuration unit 63008 according to the implementation manner have been described above in the description of the attribute encoding process of the transmitting device, so the description thereof will be skipped.

[0542] According to an embodiment, the neighbor selection related option information may include information about a basic neighbor distance calculation method (e.g., nn_base_distance_calculation_method_type), basic neighbor distance information (e.g., nn_base_distance), maximum neighbor range information (e.g., nearest_neighbore_max_range), minimum neighbor range information (e.g., nearest_neighbor_min_range), information about a method for applying the maximum neighbor distance (e.g., nn_range_filtering_location_type), information about whether to automatically calculate the maximum neighbor range (e.g., automatic_nn_range_calculation_flag), information about a method for calculating the maximum neighbor range (e.g., automatic_nn_range_method_type), and at least one of information related to the Kn value (e.g., automatic_max_nn_range_in_table, automatic_nn_range_table_k). According to an embodiment, the neighbor point selection related option information may also include information about the maximum number of points that can be set as neighbor points (e.g., lifting_num_pred_nearest_neighbours), search range related information (e.g., lifting_search_range) and / or LOD configuration method. In the present disclosure, for simplicity, at least one of the information about the basic neighbor point distance calculation method (e.g., nn_base_distance_calculation_method_type) and the basic neighbor point distance information (e.g., nn_base_distance) may be referred to as basic neighbor point distance related information.

[0543] Depending on an embodiment, information about a base neighbor distance calculation method (e.g., nn_base_distance_calculation_method_type) may indicate the use of a base distance input as a user parameter (or encoder option), an octree-based maximum neighbor distance calculation, a distance-based maximum neighbor distance calculation, a sampling-based maximum neighbor distance calculation, a maximum neighbor distance calculation by calculating the average difference of Morton codes between LODs, or a maximum neighbor distance calculation by calculating the average distance difference between LODs.

[0544] According to an embodiment, information about a method of applying a maximum neighboring point distance (e.g., nn_range_filtering_location_type) may indicate a method of calculating the distance between points when selecting neighboring points without selecting points outside the maximum neighboring point distance or a method of registering all neighboring points by calculating the distance between points and removing points exceeding the maximum neighboring point distance among the registered points.

[0545] According to an embodiment, information on whether the maximum neighboring point range is automatically calculated (eg, automatic_nn_range_calculation_flag) may indicate whether the maximum neighboring point range is automatically calculated.

[0546] According to an embodiment, information about a method of calculating a maximum neighboring point range (e.g., automatic_nn_range_method_type) may indicate a method of calculating a maximum neighboring point range by estimating density based on a diagonal length of a bounding box or a method of calculating a maximum neighboring point range by estimating density based on an axis length of a bounding box and the number of points.

[0547] The method of calculating the maximum neighboring point range by estimating the density based on the diagonal length of the bounding box may include estimating the density using the diagonal length of the bounding box as a denominator and estimating the density using the diagonal length of the bounding box as a numerator.

[0548] The method of calculating the maximum neighboring point range by estimating the density based on the axis length of the bounding box and the number of points may include a method of estimating the density based on the maximum axis length of the bounding box and the number of points, a method of estimating the density based on the minimum axis length of the bounding box and the number of points, and a method of estimating the density based on the intermediate axis length and the number of points.

[0549] According to an embodiment, the neighboring point set configuration unit 63008 estimates the density for automatic calculation using one or a combination of two or more of the following methods: a method of estimating the density using the diagonal length of the bounding box as a denominator, a method of estimating the density using the diagonal length of the bounding box as a numerator, a method of estimating the density based on the maximum axis length of the bounding box and the number of points, a method of estimating the density based on the minimum axis length of the bounding box and the number of points, and a method of estimating the density based on the intermediate axis length and the number of points.

[0550] The information related to the Kn value may include the number of entries in the table of Kn values ​​(eg, automatic_max_nn_range_in_table) and the Kn value (eg, automatic_nn_range_table_k).

[0551] According to an embodiment, the neighboring point set configuration unit 63008 may automatically calculate the maximum neighboring point range (NN_range) according to information about whether to automatically calculate the maximum neighboring point range (e.g., automatic_nn_range_calculation_flag). For example, when the information about whether to automatically calculate the maximum range (e.g., automatic_nn_range_calculation_flag) indicates automatic calculation, the maximum neighboring point range (NN_range) may be automatically calculated based on information related to the basic neighboring point distance, information about the maximum neighboring point range calculation method, and information related to the Kn value.

[0552] According to an embodiment, if the maximum neighboring point distance is determined as described above, the neighboring point set configuration unit 63008 is based on the LOD l The set is used to search for X (e.g., 3) NN points among the points within the search range in a group with the same or lower LOD (i.e., the distance between nodes is large), such as Fig.28 . Then, the neighbor point set configuration unit 63008 can register only the NN points within the maximum neighbor point distance among the X (for example, 3) NN points as the neighbor point set. Therefore, the number of NN points registered as the neighbor point set is equal to or less than X. In other words, the NN points among the X NN points that are not within the maximum neighbor point distance are not registered as the neighbor point set and are excluded from the neighbor point set. For example, if two of the three NN points are not within the maximum neighbor point distance, the two NN points are excluded and only the other NN point (i.e., one NN point) is registered as the neighbor point set.

[0553] According to another embodiment, if the maximum neighboring point distance is determined as described above, then Fig.29 In the embodiment, the neighboring point set configuration unit 53008 can search for X (for example, 3) NN points within the maximum neighboring point distance among the points within the search range in the group with the same or lower LOD (i.e., the distance between the nodes is large) based on the LOD1 set, and register the X NN points as a neighboring point set.

[0554] That is, the number of NN points that can be registered as a neighbor point set can vary depending on the timing (or position) at which the maximum neighbor point distance is applied. Fig.29 In the method, the number of NN points registered as the neighboring point set may differ depending on whether the maximum neighboring point distance is applied after searching for X NN points by neighboring point search or when the distance between points is calculated.

[0555] Refer to the example Fig.28, the neighboring point set configuration unit 63008 can compare the distance values ​​between the points in the actual search range and the point Px to search for X NN points, and only register the points within the maximum neighboring point distance at the LOD to which the point Px belongs among the X points as the neighboring point set of the point Px. That is, the neighboring points registered as the neighboring point set of the point Px are limited to the points within the maximum neighboring point distance at the LOD to which the point Px belongs among the X points.

[0556] Refer to the example Fig.29 , when comparing the distance values ​​between the points in the actual search range and the point Px, the neighboring point set configuration unit 63008 selects X (for example, 3) NN points within the maximum neighboring point distance at the LOD to which the point Px belongs from among the points in the actual search range, and registers the selected points as the neighboring point set of the point Px. That is, the neighboring points registered as the neighboring point set of the point Px are limited to the points within the maximum neighboring point distance at the LOD to which the point Px belongs.

[0557] exist Fig.28 and Fig.29 In the lifting_search_range field, the actual search range may be a value obtained by multiplying the value of the lifting_search_range field by 2 and then adding the center point to the obtained value (i.e., (the value of the lifting_search_range field × 2) + the center point), a value obtained by adding the center value to the value of the lifting_search_range field (i.e., the value of the lifting_search_range field + the center point), or the value of the lifting_search_range field. In an embodiment, the neighboring point set configuration unit 63008 searches for neighboring points of the point Px in the reserved list when the number of LODs is multiple, and searches for neighboring points of the point Px in the index list when the number of LODs is 1.

[0558] For example, if the number of LODs is 2 or more and the information about the search range (e.g., lifting_search_range field) is 128, the actual search range includes the center point Pi in the retention list arranged by the Morton code, the 128 points before the center point Pi, and the 128 points after the center point Pi. As another example, if the number of LODs is 1 and the information about the search range is 128, the actual search range includes the center point Pi in the index list arranged by the Morton code and the 128 points before the center point Pi. As another example, if the number of LODs is 1 and the information related to the search range is 128, the actual search range includes the 128 points before the current point Px in the index list arranged by the Morton code.

[0559] For example, assume that the neighboring point set configuration unit 63008 selects points P2 P4 P6 as neighboring points of point P3 (i.e., node) belonging to LOD1, and registers the selected points as a neighboring point set in the predictor of P3 (see Fig. 9 ).

[0560] According to an embodiment, when a neighboring point set is registered in each predictor of a point to be decoded in the neighboring point set configuration unit 63008, the attribute information prediction unit (attribute information predictor) 63009 predicts an attribute value of a corresponding point from one or more neighboring points registered in each predictor.

[0561] According to an embodiment, the attribute information predictor 63009 performs a process of predicting the attribute value of a point based on a prediction mode of a specific point. The attribute prediction process is performed on all points or at least some points of the reconstructed geometry.

[0562] The prediction mode of a specific point according to an embodiment may be one of prediction mode 0 to prediction mode 3.

[0563] According to an embodiment, prediction mode 0 is a mode for calculating a predicted attribute value by weighted averaging, prediction mode 1 is a mode for determining an attribute of a first neighboring point as a predicted attribute value, prediction mode 2 is a mode for determining an attribute of a second neighboring point as a predicted attribute value, and prediction mode 3 is a mode for determining an attribute of a third neighboring point as a predicted attribute value.

[0564] According to an embodiment, when the maximum difference between the attribute values ​​of neighboring points registered in the predictor of a point is less than a preset threshold, the attribute encoder on the transmitting side sets prediction mode 0 as the prediction mode of the point. When the maximum difference is greater than or equal to the preset threshold, the attribute encoder applies the RDO method to a plurality of candidate prediction modes and sets one of the candidate prediction modes as the prediction mode of the point. In an embodiment, this process is performed for each point.

[0565] According to an embodiment, the prediction mode (predIndex) of the point selected by applying the RDO method may be signaled in the attribute slice data. Accordingly, the prediction mode of the point may be obtained from the attribute slice data.

[0566] According to an embodiment, the attribute information prediction unit 63009 may predict the attribute value of each point based on the prediction mode of each point set as described above.

[0567] For example, when it is assumed that the prediction mode of point P3 is prediction mode 0, the average value of the values ​​obtained by multiplying the attributes of points P2, P4, and P6, which are neighboring points registered in the predictor of point P3, by weights (or normalized weights) may be calculated. The calculated average value may be determined as the predicted attribute value of the point.

[0568] As another example, when it is assumed that the prediction mode of the point P3 is prediction mode 1, the attribute value of the point P4 which is a neighboring point registered in the predictor of the point P3 may be determined as the predicted attribute value of the point P3.

[0569] As another example, when it is assumed that the prediction mode of the point P3 is prediction mode 2, the attribute value of the point P6 which is a neighboring point registered in the predictor of the point P3 may be determined as the predicted attribute value of the point P3.

[0570] As another example, when it is assumed that the prediction mode of the point P3 is prediction mode 3, the attribute value of the point P2 which is a neighboring point registered in the predictor of the point P3 may be determined as the predicted attribute value of the point P3.

[0571] Once the attribute information prediction unit 63009 obtains the predicted attribute value of the point based on the prediction mode of the point, the residual attribute information inverse quantization processor 63010 restores the attribute value of the point by adding the predicted attribute value of the point predicted by the predictor 63009 to the residual attribute value of the received point, and then performs inverse quantization as the inverse process of the quantization process of the sending device.

[0572] In an embodiment, when the transmitting side applies zero-run-length coding to the residual property value of a point, the residual property information inverse quantization processor 63010 performs zero-run-length decoding on the residual property value of the point and then performs inverse quantization.

[0573] The attribute value restored by the residual attribute information inverse quantization processor 63010 is output to the inverse color transformation processor 63011.

[0574] The inverse color transform processor 63011 performs inverse transform encoding of the inverse transform of the color value (or texture) included in the restored attribute value, and then outputs the attribute to the post-processor 61005. The inverse color transform processor 63011 performs the inverse transform encoding of the inverse transform of the color value (or texture) included in the restored attribute value. Fig.11 The inverse color conversion unit 11010 or Fig.13 The operation and / or inverse transform encoding of the color inverse transform processor 13010 is the same as or similar to the operation and / or inverse transform encoding.

[0575] The post-processor 61005 can reconstruct the point cloud data by matching the position restored and output by the geometry decoder 61003 with the attribute restored and output by the attribute decoder 61004. In addition, when the reconstructed point cloud data is in units of tiles and / or slices, the post-processor 61005 can perform the inverse processing of the spatial segmentation on the transmitting side based on the signaling information. Fig.16 (b) and Fig.16 As shown in (c) Fig.16When the bounding box shown in (a) is divided into tiles and slices, the tiles and / or slices can be combined based on signaling information to restore the Fig.16 The bounding box shown in (a).

[0576] Fig.32 An example of a bit stream structure for transmitted / received point cloud data according to an embodiment is illustrated.

[0577] Relevant information may be signaled to add / perform the above-mentioned embodiments. The signaling information according to the embodiments may be used in a point cloud video encoder at a transmitting end or a point cloud video decoder at a receiving end.

[0578] The point cloud video encoder according to the embodiment may generate a point cloud video by encoding the geometric information and the attribute information as described above. Fig.32 In addition, signaling information about the point cloud data may be generated and processed by at least one of a geometry encoder, an attribute encoder, or a signaling processor of the point cloud video encoder, and may be included in the bitstream.

[0579] The signaling information according to the embodiment may be received / obtained by at least one of a geometry decoder, an attribute decoder, and a signaling processor of a point cloud video decoder.

[0580] A bitstream according to an embodiment may be divided into a geometry bitstream, an attribute bitstream, and a signaling bitstream to be transmitted / received, or may be combined into one bitstream and transmitted / received.

[0581] When the geometry bitstream, attribute bitstream and signaling bitstream according to the embodiment are configured as one bitstream, the bitstream may include one or more sub-bitstreams. The bitstream according to the embodiment may include a sequence parameter set (SPS) for sequence-level signaling, a geometry parameter set (GPS) for signaling geometry information encoding, one or more attribute parameter sets (APS) (APS0, APS1) for signaling attribute information encoding, a tile parameter set (TPS) for tile-level signaling, and one or more slices (slice 0 to slice n). That is, the bitstream of point cloud data according to the embodiment may include one or more tiles, and each tile may be a group of slices including one or more slices (slice 0 to slice n). The TPS according to the embodiment may contain information about each of the one or more tiles (e.g., coordinate value information and height / size information about the bounding box). Each slice may include a geometry bitstream (Geom0) and one or more attribute bitstreams (Attr0 and Attr1). For example, the first slice (slice 0) may include a geometry bitstream (Geom0 0) and one or more attribute bitstreams (Attr 0 、Attr1 0 ).

[0582] The geometry bitstream (or geometry slice) in each slice may be composed of a geometry slice header (geom_slice_header) and geometry slice data (geom_slice_data). According to an embodiment, geom_slice_header may include identification information (geom_parameter_set_id) of a parameter set included in GPS, a tile identifier (geom_tile_id) and a slice identifier (geom_slice_id), and information (geomBoxOrigin, geom_box_log2_scale, geom_max_node_size_log2, geom_num_points) about the data contained in the geometry slice data (geom_slice_data). geomBoxOrigin is geometry box origin information indicating the origin of the box of the geometry slice data, geom_box_log2_scale is information indicating the logarithmic scaling (log scale) of the geometry slice data, geom_max_node_size_log2 is information indicating the root geometry octree node size, and geom_num_points is information related to the number of points of the geometry slice data. According to an embodiment, geom_slice_data may include geometric information (or geometric data) about point cloud data in a corresponding slice.

[0583] Each attribute bitstream (or attribute slice) in each slice may be composed of an attribute slice header (attr_slice_header) and attribute slice data (attr_slice_data). According to an embodiment, attr_slice_header may include information about the corresponding attribute slice data. The attribute slice data may contain attribute information (or attribute data or attribute value) about the point cloud data in the corresponding slice. When there are multiple attribute bitstreams in a slice, each bitstream may contain different attribute information. For example, one attribute bitstream may contain attribute information corresponding to color, while another attribute stream may contain attribute information corresponding to reflectivity.

[0584] Fig.33 An exemplary bitstream structure of point cloud data according to an embodiment is shown.

[0585] Fig.34 The connection relationship between components in a bit stream of point cloud data according to an embodiment is illustrated.

[0586] Fig.33 and Fig.34 The bitstream structure of the point cloud data illustrated in can be represented by Fig.32 The bit stream structure of the point cloud data is shown in FIG.

[0587] According to an embodiment, the SPS may include an identifier (seq_parameter_set_id) for identifying the SPS, and the GPS may include an identifier (geom_parameter_set_id) for identifying the GPS and an identifier (seq_parameter_set_id) indicating the activated SPS to which the GPS belongs. The APS may include an identifier (attr_parameter_set_id) for identifying the APS and an identifier (seq_parameter_set_id) indicating the active SPS to which the APS belongs.

[0588] According to an embodiment, the geometry bitstream (or geometry slice) may include a geometry slice header and geometry slice data. The geometry slice header may include an identifier (geom_parameter_set_id) of the active GPS referenced by the corresponding geometry slice. In addition, the geometry slice header may also include an identifier (geom_slice_id) for identifying the corresponding geometry slice and / or an identifier (geom_tile_id) for identifying the corresponding tile. The geometry slice data may include geometry information belonging to the corresponding slice.

[0589] According to an embodiment, an attribute bitstream (or attribute slice) may include an attribute slice header and attribute slice data. The attribute slice header may include an identifier (attr_parameter_set_id) of an active APS to be referenced by the corresponding attribute slice and an identifier (geom_slice_id) for identifying a geometric slice related to the attribute slice. The attribute slice data may include attribute information belonging to the corresponding slice.

[0590] That is, the geometry slice references the GPS, and the GPS references the SPS. In addition, the SPS lists the available attributes, assigns identifiers to each attribute, and identifies the decoding method. Based on the identifier, the attribute slice is mapped to the output attribute. The attribute slice depends on the previous (decoded) geometry slice and the APS. The APS references the SPS.

[0591] According to an embodiment, parameters necessary for encoding point cloud data may be newly defined in a parameter set of point cloud data and / or a corresponding slice header. For example, when encoding of attribute information is performed, parameters may be added to the APS. When tile-based encoding is performed, parameters may be added to the tile and / or slice header.

[0592] like Fig.32 , Fig.33 and Fig.34 As shown in , the present disclosure provides tiles or slices so that point cloud data can be segmented and processed by region. According to an embodiment, corresponding regions of the bitstream may have different importance. Therefore, when the point cloud data is segmented into tiles, different filters (encoding methods) and different filter units may be applied to each tile. When the point cloud data is segmented into slices, different filters and different filter units may be applied to each slice.

[0593] When point cloud data is segmented and compressed, the transmitting device and the receiving device according to the embodiment may transmit and receive a bit stream in a high-level syntax structure to selectively transmit attribute information in the segmented regions.

[0594] The transmitting device according to the embodiment may be configured as follows: Fig.32 , Fig.33 and Fig.34 The point cloud data can be transmitted using the bitstream structure shown in . Therefore, a method for applying different encoding operations to important areas and using a good quality encoding method can be provided. In addition, efficient encoding and transmission can be supported according to the characteristics of the point cloud data, and attribute values ​​can be provided according to user requirements.

[0595] The receiving device according to the embodiment may be configured as follows: Fig.32 , Fig.33 and Fig.34 The point cloud data is received by the bitstream structure shown in . Therefore, different filtering (decoding) methods can be applied to corresponding areas (areas divided into tiles or slices) instead of applying complex decoding (filtering) methods to the entire point cloud data. Therefore, better image quality in important areas is provided to the user, and appropriate latency of the system can be ensured.

[0596] As described above, tiles or slices are provided to process point cloud data by segmenting the point cloud data by region. When segmenting the point cloud data by region, an option to generate a different set of neighboring points for each region may be provided. Thus, a selection method with low complexity and slightly lower reliability or a selection method with high complexity and high reliability may be provided.

[0597] Neighboring point selection related option information of a sequence required in the process of encoding / decoding attribute information according to an embodiment may be signaled in the SPS and / or the APS.

[0598] According to an embodiment, if there are tiles or slices with different attribute characteristics in the same sequence, the neighbor point selection related option information of the sequence may be signaled in the TPS and / or attribute slice header of each slice.

[0599] According to an embodiment, when the point cloud data is divided into regions, the attribute characteristics of a specific region may be different from the attribute characteristics of a sequence, and thus may be configured differently through different maximum neighboring point range configuration functions.

[0600] Therefore, when the point cloud data is divided into tiles, a different maximum neighboring point range can be applied to each tile. In addition, when the point cloud data is divided into slices, a different maximum neighboring point range can be applied to each slice.

[0601] At least one of the SPS, APS, TPS and attribute slice header of each slice according to an embodiment may include neighboring point selection related option information. According to an embodiment, the neighboring point selection related option information may include information related to a maximum neighboring point range.

[0602] According to an embodiment, the neighbor selection related option information may include information about a basic neighbor distance calculation method (e.g., nn_base_distance_calculation_method_type), basic neighbor distance information (e.g., nn_base_distance), maximum neighbor range information (e.g., nearest_neighbore_max_range), minimum neighbor range information (e.g., nearest_neighbor_min_range), information about a method for applying the maximum neighbor distance (e.g., nn_range_filtering_location_type), information about whether to automatically calculate the maximum neighbor range (e.g., automatic_nn_range_calculation_flag), information about a method for calculating the maximum neighbor range (e.g., automatic_nn_range_method_type), and at least one of information related to the Kn value (e.g., automatic_max_nn_range_in_table, automatic_nn_range_table_k).

[0603] According to an embodiment, the neighboring point selection related option information may further include information about the maximum number of points that can be set as neighboring points (eg, lifting_num_pred_nearest_neighbours), search range related information (eg, lifting_search_range), and / or an LOD configuration method.

[0604] A field, which is a term used in the syntax of the present disclosure described later, may have the same meaning as a parameter or an element.

[0605] Fig.35 An embodiment of the syntax structure of a sequence parameter set (SPS) (seq_parameter_set_rbsp()) according to the present disclosure is shown. The SPS may include sequence information about a point cloud data bitstream. In particular, in this example, the SPS includes neighbor point selection related option information.

[0606] The SPS according to an embodiment may include a profile_idc field, a profile_compatibility_flags field, a level_idc field, a sps_bound_box_present_flag field, a sps_source_scale_factor field, a sps_seq_parameter_set_id field, a sps_num_attribute_sets field, and a sps_extension_present_flag field.

[0607] The profile_idc field indicates the profile to which the bitstream conforms.

[0608] The profile_compatibility_flags field equal to 1 may indicate that the bitstream conforms to the profile indicated by profile_idc.

[0609] The level_idc field indicates the level to which the bitstream conforms.

[0610] The sps_bounding_box_present_flag field indicates whether source bounding box information is signaled in the SPS. The source bounding box information may include offset and size information about the source bounding box. For example, an sps_bounding_box_present_flag field equal to 1 indicates that source bounding box information is signaled in the SPS. An sps_bounding_box_present_flag field equal to 0 indicates that source bounding box information is not signaled. The sps_source_scale_factor field indicates the scaling factor of the source point cloud.

[0611] The sps_seq_parameter_set_id field provides an identifier of the SPS for reference by other syntax elements.

[0612] The sps_num_attribute_sets field indicates the number of coded attributes in the bitstream.

[0613] The sps_extension_present_flag field specifies whether the sps_extension_data syntax structure is present in the SPS syntax structure. For example, a sps_extension_present_flag field equal to 1 specifies that the sps_extension_data syntax structure is present in the SPS syntax structure. A sps_extension_present_flag field equal to 0 specifies that the syntax structure is not present. When not present, the value of the sps_extension_present_flag field is inferred to be equal to 0.

[0614] When the sps_bounding_box_present_flag field is equal to 1, the SPS according to an embodiment may further include a sps_bounding_box_offset_x field, a sps_bounding_box_offset_y field, a sps_bounding_box_offset_z field, a sps_bounding_box_scale_factor field, a sps_bounding_box_size_width field, a sps_bounding_box_size_height field, and a sps_bounding_box_size_depth field.

[0615] The sps_bounding_box_offset_x field indicates the x offset of the source bounding box in Cartesian coordinates. When the x offset of the source bounding box does not exist, the value of sps_bounding_box_offset_x is 0.

[0616] The sps_bounding_box_offset_y field indicates the y offset of the source bounding box in Cartesian coordinates. When the y offset of the source bounding box does not exist, the value of sps_bounding_box_offset_y is 0.

[0617] The sps_bounding_box_offset_z field indicates the z offset of the source bounding box in Cartesian coordinates. When the z offset of the source bounding box does not exist, the value of sps_bounding_box_offset_z is 0.

[0618] The sps_bounding_box_scale_factor field indicates the scaling factor of the source bounding box in Cartesian coordinates. When the scaling factor of the source bounding box does not exist, the value of sps_bounding_box_scale_factor may be 1.

[0619] The sps_bounding_box_size_width field indicates the width of the source bounding box in Cartesian coordinates. When the width of the source bounding box does not exist, the value of the sps_bounding_box_size_width field may be 1.

[0620] The sps_bounding_box_size_height field indicates the height of the source bounding box in Cartesian coordinates. When the height of the source bounding box does not exist, the value of the sps_bounding_box_size_height field may be 1.

[0621] The sps_bounding_box_size_depth field indicates the depth of the source bounding box in Cartesian coordinates. When the depth of the source bounding box does not exist, the value of the sps_bounding_box_size_depth field may be 1.

[0622] An SPS according to an embodiment includes an iteration statement that is repeated as many times as the value of the sps_num_attribute_sets field. In an embodiment, i is initialized to 0 and incremented by 1 each time the iteration statement is executed. The iteration statement is repeated until the value of i becomes equal to the value of the sps_num_attribute_sets field. The iteration statement may include an attribute_dimension[i] field, an attribute_instance_id[i] field, an attribute_bitdepth[i] field, an attribute_cicp_colour_primaries[i] field, an attribute_cicp_transfer_characteristics[i] field, an attribute_cicp_matrix_coeffs[i] field, an attribute_cicp_video_full_range_flag[i] field, and a known_attribute_label_flag[i] field.

[0623] The attribute_dimension[i] field specifies the number of components of the i-th attribute.

[0624] The attribute_instance_id[i] field specifies the instance ID of the i-th attribute.

[0625] The attribute_bitdepth[i] field specifies the bit depth of the i-th attribute signal.

[0626] The attribute_cicp_colour_primaries[i] field indicates the chromaticity coordinates of the colour attribute source primaries of the i-th attribute.

[0627] The attribute_cicp_transfer_characteristics[i] field indicates either a reference electro-optical transfer characteristic function of the color attribute as a function of source input linear optical intensity with a nominal true value range of 0 to 1, or the inverse of the reference electro-optical transfer characteristic function as a function of output linear optical intensity.

[0628] The attribute_cicp_matrix_coeffs[i] field describes the matrix coefficients used when deriving the luma and chroma signals from the green, blue, and red or Y, Z, and X primaries.

[0629] The attribute_cicp_video_full_range_flag[i] field indicates the black level and range of luminance and chrominance signals derived from E′Y, E′PB and E′PR or E′R, E′G and E′B real-valued component signals.

[0630] The known_attribute_label_flag[i] field specifies whether the known_attribute_label field or the attribute_label_four_bytes field is signaled for the i-th attribute. For example, a value of the known_attribute_label_flag[i] field equal to 0 specifies that the known_attribute_label field is signaled for the i-th attribute. A known_attribute_label_flag[i] field equal to 1 specifies that the attribute_label_four_bytes field is signaled for the i-th attribute.

[0631] The known_attribute_label[i] field may specify the attribute type. For example, a known_attribute_label[i] field equal to 0 may specify that the i-th attribute is color. A known_attribute_label[i] field equal to 1 may specify that the i-th attribute is reflectivity. A known_attribute_label[i] field equal to 2 may specify that the i-th attribute is a frame index.

[0632] The attribute_label_four_bytes field indicates a known attribute type with a 4-byte code.

[0633] In this example, the attribute_label_four_bytes field indicates color when equal to 0 and indicates reflectivity when equal to 1.

[0634] According to an embodiment, when the sps_extension_present_flag field is equal to 1, the SPS may further include a sps_extension_data_flag field.

[0635] The sps_extension_data_flag field can have any value.

[0636] According to an implementation manner, the SPS may further include the following neighbor point selection related option information.

[0637] According to an embodiment, the neighbor point selection related option information may be included in an iteration statement having the same number of iterations as the value of the above-mentioned sps_num_attribute_sets field.

[0638] That is, the iteration statement may further include a nn_base_distance_calculation_method_type[i] field, a nearest_neighbour_max_range[i] field, a nearest_neighbour_min_range[i] field, a nn_range_filtering_location_type[i] field, and an automatic_nnn_range_calculation_flag[i] field.

[0639] The nn_base_distance_calculation_method_type[i] field may indicate the base neighbor distance calculation method applied when compressing the i-th attribute of the corresponding sequence. For example, a nn_base_distance_calculation_method_type[i] field equal to 0 may indicate the use of a base distance. A nn_base_distance_calculation_method_type[i] field equal to 1 may indicate an octree-based maximum neighbor distance calculation. A nn_base_distance_calculation_method_type[i] field equal to 2 may indicate a distance-based maximum neighbor distance calculation. A nn_base_distance_calculation_method_type[i] field equal to 3 may indicate a sampling-based maximum neighbor distance calculation. A nn_base_distance_calculation_method_type[i] field equal to 4 may indicate a maximum neighbor distance calculation performed by calculating the average difference of the Morton codes between the LODs. A nn_base_distance_calculation_method_type[i] field equal to 5 may indicate a maximum neighbor distance calculation performed by calculating the average distance difference between the LODs.

[0640] According to an embodiment, when the value of the nn_base_distance_calculation_method_type[i] field is 0, that is, when the field indicates the use of the input base distance, the iteration statement may further include the nn_base_distance[i] field.

[0641] The nn_base_distance[i] field may indicate a base neighbor distance applied when compressing the i-th attribute of the corresponding sequence.

[0642] The nearest_neighbour_max_range[i] field may indicate the maximum neighboring point range applied when compressing the i-th attribute of the sequence. According to an embodiment, the value of the nearest_neighbour_max_range[i] field may be used as the value of NN_range in Formula 5. According to an embodiment, the distance of the point registered as a neighboring point may be limited using the nearest_neighbour_max_range[i] field. For example, when the LOD is generated based on an octree, the value of the nearest_neighbour_max_range[i] field may be the number of octree nodes surrounding the point.

[0643] The nearest_neighbour_min_range[i] field may indicate the minimum neighbor range applied when compressing the i-th attribute of the sequence.

[0644] The nn_range_filtering_location_type[i] field may indicate a method of applying a maximum neighboring point distance when compressing the i-th attribute of a sequence. For example, when the value of the nn_range_filtering_location_type[i] field is 0, it may indicate a method of calculating the distance between points when selecting neighboring points and not selecting points outside the maximum neighboring point distance. As another example, when the value of the nn_range_filtering_location_type[i] field is 1, it may indicate a method of registering all neighboring points by calculating the distance between points and removing points exceeding the maximum neighboring point distance among the registered points.

[0645] The automatic_nn_range_calculation_flag[i] field may indicate whether a maximum neighbor range is automatically calculated when compressing the i-th attribute of a sequence. For example, when the value of the automatic_nn_range_calculation_flag[i] field is 0 (ie, TRUE), it may indicate automatic calculation of the maximum neighbor range.

[0646] According to an embodiment, when the value of the automatic_nn_range_calculation_flag[i] field is 0, that is, when the field indicates automatic calculation of the maximum neighboring point range, the iteration statement may further include an automatic_nn_range_method_type[i] field and an automatic_max_nn_range_in_table[i] field.

[0647] The automatic_nn_range_method_type[i] field may indicate a method of calculating a maximum neighboring point range when compressing the i-th attribute of a sequence.

[0648] For example, an automatic_nn_range_method_type[i] field equal to 1 may indicate a method of estimating density using the diagonal length of the bounding box as the denominator. An automatic_nn_range_method_type[i] field equal to 2 may indicate a method of estimating density using the diagonal length of the bounding box as the numerator. An automatic_nn_range_method_type[i] field equal to 3 may indicate a method of estimating density based on the maximum axis length and the number of points of the bounding box. An automatic_nn_range_method_type[i] field equal to 4 may indicate a method of estimating density based on the minimum axis length and the number of points of the bounding box. An automatic_nn_range_method_type[i] field equal to 5 may indicate a method of estimating density based on the intermediate axis length and the number of points. As another example, an automatic_nn_range_method_type[i] field equal to 6 may indicate a calculation method that combines a method of estimating density using the diagonal length of the bounding box as the denominator with a method of estimating density based on the maximum axis length and the number of points of the bounding box. The automatic_nn_range_method_type[i] field equal to 7 may indicate a calculation method that combines a method of estimating density using the diagonal length of a bounding box as a denominator with a method of estimating density based on the minimum axis length of the bounding box and the number of points. The automatic_nn_range_method_type[i] field equal to 8 may indicate a calculation method that combines a method of estimating density using the diagonal length of a bounding box as a denominator with a method of estimating density based on the median axis length of the bounding box and the number of points. The automatic_nn_range_method_type[i] field equal to 9 may indicate a calculation method that combines a method of estimating density using the diagonal length of a bounding box as a numerator with a method of estimating density based on the maximum axis length of the bounding box and the number of points. The automatic_nn_range_method_type[i] field equal to 10 may indicate a calculation method that combines a method of estimating density using the diagonal length of a bounding box as a numerator with a method of estimating density based on the minimum axis length of the bounding box and the number of points. The automatic_nn_range_method_type[i] field equal to 11 may indicate a method of estimating density using the diagonal length of the bounding box as a numerator and a method of estimating density based on the medial axis length of the bounding box and the number of points.

[0649] The automatic_max_nn_range_in_table[i] field may indicate the number of entries in the table of Kn values ​​when compressing the i-th attribute of a sequence.

[0650] An SPS according to an embodiment may include an iteration statement with as many iterations as the value of the automatic_max_nn_range_in_table[i] field. In this case, according to an embodiment, j is initialized to 0 and incremented by 1 each time an iteration statement is executed. The iteration statement is iterated until the j value becomes the value of the automatic_max_nn_range_in_table[i] field. The iteration statement may include an automatic_nn_range_table_k[i][j] field.

[0651] The automatic_nn_range_table_k[i][j] field may indicate a Kn value when compressing the i-th attribute of the corresponding sequence.

[0652] Fig.36 An embodiment of a syntax structure of a geometry parameter set (GPS) (seq_parameter_set_rbsp()) according to the present disclosure is shown. The GPS according to an embodiment may include information on a method of encoding geometry information in point cloud data included in one or more slices.

[0653] According to an embodiment, the GPS may include a gps_geom_parameter_set_id field, a gps_seq_parameter_set_id field, a gps_box_present_flag field, a unique_geometry_points_flag field, a neighbor_context_restriction_flag field, an inferred_direct_coding_mode_enabled_flag field, a bitwise_occupancy_coding_flag field, an adjacent_child_contextualization_enabled_flag field, a log2_neighbour_avail_boundary field, a log2_intra_pred_max_node_size field, a log2_trisoup_node_size field, and a gps_extension_present_flag field.

[0654] The gps_geom_parameter_set_id field provides an identifier of the GPS for reference by other syntax elements.

[0655] The gps_seq_parameter_set_id field specifies the value of sps_seq_parameter_set_id for the active SPS.

[0656] The gps_box_present_flag field specifies whether additional bounding box information is provided in the geometry slice header that references the current GPS. For example, a gps_box_present_flag field equal to 1 may specify that additional bounding box information is provided in the geometry header that references the current GPS. Thus, when the gps_box_present_flag field is equal to 1, the GPS may also include a gps_gsh_box_log2_scale_present_flag field.

[0657] The gps_gsh_box_log2_scale_present_flag field specifies whether the gps_gsh_box_log2_scale field is signaled in each geometry slice header that references the current GPS. For example, a gps_gsh_box_log2_scale_present_flag field equal to 1 may specify that the gps_gsh_box_log2_scale field is signaled in each geometry slice header that references the current GPS. As another example, a gps_gsh_box_log2_scale_present_flag field equal to 0 may specify that the gps_gsh_box_log2_scale field is not signaled in each geometry slice header and that a common scaling of all slices is signaled in the gps_gsh_box_log2_scale field of the current GPS.

[0658] When the gps_gsh_box_log2_scale_present_flag field is equal to 0, GPS may also include a gps_gsh_box_log2_scale field.

[0659] The gps_gsh_box_log2_scale field indicates the common scaling factor of the bounding box origin of all slices referenced to the current GPS.

[0660] The unique_geometry_points_flag field indicates whether all output points have unique locations. For example, a unique_geometry_points_flag field equal to 1 indicates that all output points have unique locations. A unique_geometry_points_flag field equal to 0 indicates that two or more of the output points may have the same location in all slices referenced to the current GPS.

[0661] The neighbor_context_restriction_flag field indicates the context used for octree occupancy coding. For example, a neighbor_context_restriction_flag field equal to 0 indicates that the octree occupancy coding uses the context determined from the six neighboring parent nodes. A neighbor_context_restriction_flag field equal to 1 indicates that the octree occupancy coding uses only the context determined from the sibling nodes.

[0662] The inferred_direct_coding_mode_enabled_flag field indicates whether a direct_mode_flag field is present in the geometry node syntax. For example, an inferred_direct_coding_mode_enabled_flag field equal to 1 indicates that a direct_mode_flag field may be present in the geometry node syntax. For example, an inferred_direct_coding_mode_enabled_flag field equal to 0 indicates that a direct_mode_flag field is not present in the geometry node syntax.

[0663] The bitwise_occupancy_coding_flag field indicates whether the bitwise contextualization of the syntax element occupancy map is used to encode the geometry node occupancy. For example, a bitwise_occupancy_coding_flag field equal to 1 indicates that the bitwise contextualization of the syntax element ocupancy_map is used to encode the geometry node occupancy. For example, a bitwise_occupancy_coding_flag field equal to 0 indicates that the dictionary-coded syntax element occupancy_byte is used to encode the geometry node occupancy.

[0664] The adjacent_child_contextualization_enabled_flag field indicates whether the adjacent children of the adjacent octree node are used for bitwise occupancy contextualization. For example, an adjacent_child_contextualization_enabled_flag field equal to 1 indicates that the adjacent children of the adjacent octree node are used for bitwise occupancy contextualization. For example, an adjacent_child_contextualization_enabled_flag field equal to 0 indicates that the children of the adjacent octree node are not used for occupancy contextualization.

[0665] The log2_neighbour_avail_boundary field specifies the value of the variable NeighbAvailBoundary used in the decoding process as follows:

[0666] NeighbAvailBoundary=2 log2_neighbour_avail_boundary .

[0667] For example, when the neighbor_context_restriction_flag field is equal to 1, the NeighbAvailabilityMask may be set equal to 1. For example, when the neighbor_context_restriction_flag field is equal to 0, the NeighbAvailabilityMask may be set equal to 1. <log2_neighbour_avail_boundary。

[0668] The log2_intra_pred_max_node_size field specifies the size of octree nodes that are eligible to occupy intra prediction.

[0669] The log2_trisoup_node_size field specifies the variable TrisoupNodeSize as the size of the triangle nodes as follows:

[0670] TrisoupNodeSize=1< <log2_trisoup_node_size。

[0671] The gps_extension_present_flag field specifies whether a gps_extension_data syntax structure is present in the GPS syntax structure. For example, a gps_extension_present_flag equal to 1 specifies that a gps_extension_data syntax structure is present in the GPS syntax. For example, a gps_extension_present_flag equal to 0 specifies that the syntax structure is not present in the GPS syntax.

[0672] When the value of the gps_extension_present_flag field is equal to 1, the GPS according to an embodiment may further include a gps_extension_data_flag field.

[0673] The gps_extension_data_flag field can have any value. Its presence and value do not affect the conformance of the decoder to the profile.

[0674] Fig.37 An embodiment of a syntax structure of an attribute parameter set (APS) (attribute_parameter_set()) according to the present disclosure is shown. An APS according to an embodiment may include information about a method for encoding attribute information in point cloud data included in one or more slices. According to an embodiment, an APS may include neighbor point selection related option information.

[0675] The APS according to an embodiment may include an aps_attr_parameter_set_id field, an aps_seq_parameter_set_id field, an attr_coding_type field, an aps_attr_initial_qp field, an aps_attr_chroma_qp_offset field, an aps_slice_qp_delta_present_flag field, and an aps_flag field.

[0676] The aps_attr_parameter_set_id field provides an identifier of the APS for reference by other syntax elements.

[0677] The aps_seq_parameter_set_id field indicates the value of sps_seq_parameter_set_id of the active SPS.

[0678] The attr_coding_type field indicates the coding type of the attribute.

[0679] In this example, an attr_coding_type field equal to 0 indicates prediction weight boosting (or LOD with prediction transform) as the encoding type. An attr_coding_type field equal to 1 indicates RAHT as the encoding type. An attr_coding_type field equal to 2 indicates fixed weight boosting (or LOD with lifting transform).

[0680] The aps_attr_initial_qp field indicates the initial value of the variable SliceQp for each slice that references the APS. The initial value of SliceQp is modified at the attribute slice segment level when a non-zero value of slice_qp_delta_luma or slice_qp_delta_luma is decoded.

[0681] The aps_attr_chroma_qp_offset field indicates an offset to the initial quantization parameter signaled by the syntax aps_attr_initial_qp.

[0682] The aps_slice_qp_delta_present_flag field indicates whether the ash_attr_qp_delta_luma and ash_attr_qp_delta_chroma syntax elements are present in the attribute slice header (ASH). For example, an aps_slice_qp_delta_present_flag field equal to 1 indicates that the ash_attr_qp_delta_luma and ash_attr_qp_delta_chroma syntax elements are present in the ASH. For example, an aps_slice_qp_delta_present_flag field equal to 0 indicates that the ash_attr_qp_delta_luma and ash_attr_qp_delta_chroma syntax elements are not present in the ASH.

[0683] When the value of the attr_coding_type field is 0 or 2, that is, when the encoding type is prediction weight lifting (or LOD with prediction transform) or fixed weight lifting (or LOD with lifting transform), the APS according to the embodiment may also include a lifting_num_pred_nearest_neighbours field, a lifting_max_num_direct_predictors field, a lifting_search_range field, a lifting_lod_regular_sampling_enabled_flag field, and a lifting_num_detail_levels_minus1 field.

[0684] The lifting_num_pred_nearest_neighbours field indicates the maximum number of nearest neighbors (ie, X values) to be used for prediction.

[0685] The lifting_max_num_direct_predictors field indicates the maximum number of predictors to be used for direct prediction. The value of the variable MaxNumPredictors used in the decoding process is as follows:

[0686] MaxNumPredictors=lifting_max_num_direct_predictors field+1

[0687] The lifting_search_range field indicates the search range used to determine the nearest neighbor.

[0688] The lifting_num_detail_levels_minus1 field indicates the number of detail levels that the attribute encodes.

[0689] The lifting_lod_regular_sampling_enabled_flag field indicates whether the level of detail (LOD) is constructed by using a regular sampling strategy. For example, lifting_lod_regular_sampling_enabled_flag equal to 1 indicates that the level of detail (LOD) is constructed by using a regular sampling strategy. Lifting_lod_regular_sampling_enabled_flag equal to 0 indicates that a distance-based sampling strategy is used instead.

[0690] The APS according to an embodiment includes an iteration statement that is repeated as many times as the value of the lifting_num_detail_levels_minus1 field. In an embodiment, the index (idx) is initialized to 0 and incremented by 1 each time the iteration statement is executed, and the iteration statement is repeated until the index (idx) is greater than the value of the lifting_num_detail_levels_minus1 field. When the value of the lifting_lod_decimation_enabled_flag field is true (e.g., 1), the iteration statement may include the lifting_sampling_period[idx] field, and when the value of the lifting_lod_decimation_enabled_flag field is false (e.g., 0), the iteration statement may include the lifting_sampling_distance_squared[idx] field.

[0691] The lifting_sampling_period[idx] field indicates the sampling period of the detail level idx.

[0692] The lifting_sampling_distance_squared[idx] field indicates the square of the sampling distance of the detail level idx.

[0693] When the value of the attr_coding_type field is 0, that is, when the encoding type is prediction weight lifting (or LOD with prediction transform), the APS according to an embodiment may further include a lifting_adaptive_prediction_threshold field and a lifting_intra_lod_prediction_num_layers field.

[0694] The lifting_adaptive_prediction_threshold field indicates a threshold for enabling adaptive prediction.

[0695] The lifting_intra_lod_prediction_num_layers field indicates the number of LOD layers that can reference decoded points in the same LOD layer to generate a prediction value for a target point. For example, a lifting_intra_lod_prediction_num_layers field equal to num_detail_levels_minus1 plus 1 indicates that for all LOD layers, the target point can reference decoded points in the same LOD layer. For example, a lifting_intra_lod_prediction_num_layers field equal to 0 indicates that for any LOD layer, the target point cannot reference decoded points in the same LoD layer.

[0696] The aps_extension_present_flag field indicates whether an aps_extension_data syntax structure exists in the APS syntax structure. For example, an aps_extension_present_flag field equal to 1 indicates that an aps_extension_data syntax structure exists in the APS syntax structure. For example, an aps_extension_present_flag field equal to 0 indicates that the syntax structure does not exist in the APS syntax structure.

[0697] When a value of the aps_extension_present_flag field is 1, the APS according to an embodiment may further include an aps_extension_data_flag field.

[0698] The aps_extension_data_flag field can have any value. Its presence and value do not affect the conformance of the decoder to the profile.

[0699] The APS according to the embodiment may further include the following neighbor point selection related option information.

[0700] When the value of the attr_coding_type field is 0 or 2, that is, when the encoding type is "prediction weight lifting (or LOD with prediction transform)" or "fixed weight lifting (or LOD with lifting transform)", the APS according to an embodiment may also include a different_nn_range_in_tile_flag field and a different_nn_range_per_lod_flag field.

[0701] The different_nn_range_in_tile_flag field may indicate whether the sequence is using different maximum / minimum neighbor ranges for tiles partitioned from the sequence.

[0702] The different_nn_range_per_lod_flag field may indicate whether to use different maximum / minimum neighboring point ranges for each LOD.

[0703] For example, when the value of the different_nn_range_per_lod_flag field is FALSE, the APS may further include a nearest_neighbour_max_range field, a nearest_neighbour_min_range field, a nn_base_distance_calculation_method_type field, a nn_range_filtering_location_type field, and an automatic_nn_range_calculation_flag field.

[0704] The nearest_neighbour_max_range field may indicate the maximum neighboring point range when compressing the corresponding attribute. According to an embodiment, the distance of the point registered as a neighboring point may be limited using the nearest_neighbour_max_range field. According to an embodiment, the value of the nearest_neighbour_max_range field may be used as the value of NN_range in Formula 5.

[0705] The nearest_neighbour_min_range field may indicate the minimum neighbor range when compressing attributes.

[0706] The nn_base_distance_calculation_method_type field may indicate the base neighbor distance calculation method applied when compressing attributes.

[0707] For example, a nn_base_distance_calculation_method_type field equal to 0 may indicate the use of a base distance. A nn_base_distance_calculation_method_type field equal to 1 may indicate an octree-based maximum neighbor distance calculation. A nn_base_distance_calculation_method_type field equal to 2 may indicate a distance-based maximum neighbor distance calculation. A nn_base_distance_calculation_method_type field equal to 3 may indicate a sampling-based maximum neighbor distance calculation. A nn_base_distance_calculation_method_type field equal to 4 may indicate a maximum neighbor distance calculation by calculating the average difference of the Morton codes between the LODs. A nn_base_distance_calculation_method_type field equal to 5 may indicate a maximum neighbor distance calculation by calculating the average distance difference between the LODs.

[0708] According to an embodiment, when a value of the nn_base_distance_calculation_method_type field is 0, that is, when the field indicates the use of an input base distance, the iteration statement may further include the nn_base_distance field.

[0709] The nn_base_distance field may indicate the base neighbor distance applied when compressing attributes.

[0710] The nn_range_filtering_location_type field may indicate a method of applying a maximum neighboring point distance when compressing a corresponding attribute. For example, when the value of the nn_range_filtering_location_type field is 0, it may indicate a method of calculating the distance between points when selecting neighboring points and not selecting points outside the maximum neighboring point distance. As another example, when the value of the nn_range_filtering_location_type field is 1, it may indicate a method of registering all neighboring points by calculating the distance between points and removing points exceeding the maximum neighboring point distance among the registered points.

[0711] The automatic_nn_range_calculation_flag field may indicate whether to automatically calculate a maximum neighbor range when compressing a corresponding attribute. For example, when a value of the automatic_nn_range_calculation_flag field is 0 (ie, TRUE), it may indicate automatic calculation of a maximum neighbor range.

[0712] According to an embodiment, when the value of the automatic_nn_range_calculation_flag field is 0, that is, when the field indicates automatic calculation of the maximum neighboring point range, the iteration statement may further include an automatic_nn_range_method_type field and an automatic_max_nn_range_in_table field.

[0713] The automatic_nn_range_method_type field may indicate a method of calculating a maximum neighboring point range when compressing a corresponding attribute.

[0714] For example, an automatic_nn_range_method_type field equal to 1 may indicate a method of estimating density using the diagonal length of the bounding box as a denominator. An automatic_nn_range_method_type field equal to 2 may indicate a method of estimating density using the diagonal length of the bounding box as a numerator. An automatic_nn_range_method_type field equal to 3 may indicate a method of estimating density based on the maximum axis length and the number of points of the bounding box. An automatic_nn_range_method_type field equal to 4 may indicate a method of estimating density based on the minimum axis length and the number of points of the bounding box. An automatic_nn_range_method_type field equal to 5 may indicate a method of estimating density based on the intermediate axis length and the number of points. As another example, an automatic_nn_range_method_type field equal to 6 may indicate a calculation method that combines a method of estimating density using the diagonal length of the bounding box as a denominator with a method of estimating density based on the maximum axis length and the number of points of the bounding box. The automatic_nn_range_method_type field equal to 7 may indicate a calculation method that combines a method of estimating density using the diagonal length of a bounding box as a denominator with a method of estimating density based on the minimum axis length of the bounding box and the number of points. The automatic_nn_range_method_type field equal to 8 may indicate a calculation method that combines a method of estimating density using the diagonal length of a bounding box as a denominator with a method of estimating density based on the median axis length of the bounding box and the number of points. The automatic_nn_range_method_type field equal to 9 may indicate a calculation method that combines a method of estimating density using the diagonal length of a bounding box as a numerator with a method of estimating density based on the maximum axis length of the bounding box and the number of points. The automatic_nn_range_method_type field equal to 10 may indicate a calculation method that combines a method of estimating density using the diagonal length of a bounding box as a numerator with a method of estimating density based on the minimum axis length of the bounding box and the number of points. The automatic_nn_range_method_type field equal to 11 may indicate a method of estimating density using a diagonal length of a bounding box as a numerator and a method of estimating density based on a median axis length of a bounding box and the number of points.

[0715] The automatic_max_nn_range_in_table field may indicate the number of entries in the table of Kn values ​​when compressing the i-th attribute of a sequence.

[0716] An APS according to an embodiment may include an iteration statement with as many iterations as the value of the automatic_max_nn_range_in_table field. In this case, according to an embodiment, j is initialized to 0 and incremented by 1 each time the iteration statement is executed. The iteration statement is iterated until the j value becomes the value of the automatic_max_nn_range_in_table field. The iteration statement may include the automatic_nn_range_table_k[j] field.

[0717] The automatic_nn_range_table_k[j] field may indicate a Kn value when compressing the i-th attribute of the corresponding sequence.

[0718] According to an embodiment, when the value of the different_nn_range_per_lod_flag field is TRUE, the APS also includes an iteration statement with the same number of iterations as the value of the lifting_num_detail_levels_minus1 field. In this case, according to an embodiment, the index (idx) is initialized to 0 and incremented by 1 each time the iteration statement is executed. Iterate the iteration statement until the index (idx) is greater than the value of the lifting_num_detail_levels_minus1 field. The iteration statement may also include a nearest_neighbour_max_range[idx] field, a nearest_neighbour_min_range[idx] field, a nn_base_distance_calculation_method_type[idx] field, a nn_range_filtering_location_type[idx] field, and an automatic_nn_range_calculation_flag[idx] field.

[0719] The nearest_neighbour_max_range[idx] field may indicate the maximum neighboring point range for LOD idx. According to an embodiment, the distance of a point registered as a neighboring point of LOD idx may be limited using the nearest_neighbour_max_range[idx] field. According to an embodiment, the value of the nearest_neighbour_max_range[idx] field may be used as the value of NN_range of LOD idx.

[0720] The nearest_neighbour_min_range[idx] field may indicate the minimum neighbor range of LOD idx.

[0721] The nn_base_distance_calculation_method_type[idx] field may indicate a base neighbor distance calculation method for LOD idx. As for a value assigned to the nn_base_distance_calculation_method_type[idx] field and a definition (or meaning) of the value, refer to the description of the nn_base_distance_calculation_method_type field.

[0722] According to an embodiment, when a value of the nn_base_distance_calculation_method_type[idx] field is 0, that is, when the field indicates the use of an input base distance, the iteration statement may further include a nn_base_distance[idx] field.

[0723] The nn_base_distance[idx] field may indicate the base neighbor distance for LOD idx.

[0724] The nn_range_filtering_location_type[idx] field may indicate a method of applying a maximum neighboring point distance for LOD idx. For example, when the value of the nn_range_filtering_location_type[idx] field is 0, it may indicate a method of calculating the distance between points when selecting neighboring points and not selecting points outside the maximum neighboring point distance. As another example, when the value of the nn_range_filtering_location_type[idx] field is 1, it may indicate a method of registering all neighboring points by calculating the distance between points and removing points exceeding the maximum neighboring point distance among the registered points.

[0725] The automatic_nn_range_calculation_flag[idx] field may indicate whether to automatically calculate a maximum neighboring point range for LOD idx. For example, when a value of the automatic_nn_range_calculation_flag[idx] field is 0 (ie, TRUE), it may indicate automatic calculation of a maximum neighboring point range.

[0726] According to an embodiment, when the value of the automatic_nn_range_calculation_flag[i] field is 0, that is, when the field indicates automatic calculation of the maximum neighboring point range, the iteration statement may further include an automatic_nn_range_method_type[idx] field and an automatic_max_nn_range_in_table[idx] field.

[0727] The automatic_nn_range_method_type[idx] field may indicate a method for calculating a maximum neighboring point range when compressing a corresponding attribute. As for the value assigned to the automatic_nn_range_method_type[idx] field and the definition (or meaning) of the value, refer to the description of the value, refer to the description of the automatic_nn_range_method_type field.

[0728] The automatic_max_nn_range_in_table[idx] field may indicate the number of entries in the table of Kn values ​​for LOD idx.

[0729] An APS according to an embodiment may include an iteration statement with as many iterations as the value of the automatic_max_nn_range_in_table[idx] field. In this case, according to an embodiment, j is initialized to 0 and incremented by 1 each time the iteration statement is executed. The iteration statement is iterated until the j value becomes the value of the automatic_max_nn_range_in_table[idx] field. The iteration statement may include the automatic_nn_range_table_k[idx][j] field.

[0730] The automatic_nn_range_table_k[idx][j] field may indicate a value of Kn for LOD idx.

[0731] Fig.38 An embodiment of a grammatical structure of a tile parameter set (TPS) (tile_parameter_set()) according to the present disclosure is shown. According to an embodiment, a TPS may be referred to as a tile list. The TPS according to an embodiment includes information related to each tile. In particular, in this example, the TPS includes neighbor point selection related option information.

[0732] The TPS according to an embodiment includes a num_tiles field.

[0733] The num_tiles field indicates the number of tiles signaled for the bitstream. When not present, num_tiles is inferred to be 0.

[0734] The TPS according to an embodiment includes an iteration statement that is repeated as many times as the value of the num_tiles field. In an embodiment, i is initialized to 0 and incremented by 1 each time the iteration statement is executed. The iteration statement is repeated until the value of i becomes equal to the value of the num_tiles field. The iteration statement may include a tile_bounding_box_offset_x[i] field, a tile_bounding_box_offset_y[i] field, a tile_bounding_box_offset_z[i] field, a tile_bounding_box_size_width[i] field, a tile_bounding_box_size_height[i], and a tile_bounding_box_size_depth[i] field.

[0735] The tile_bounding_box_offset_x[i] field indicates the x offset of the i-th tile in Cartesian coordinates.

[0736] The tile_bounding_box_offset_y[i] field indicates the y offset of the i-th tile in Cartesian coordinates.

[0737] The tile_bounding_box_offset_z[i] field indicates the z offset of the i-th tile in Cartesian coordinates.

[0738] The tile_bounding_box_size_width[i] field indicates the width of the i-th tile in Cartesian coordinates.

[0739] The tile_bounding_box_size_height[i] field indicates the height of the i-th tile in Cartesian ...

Claims

1. A method for sending point cloud data, the method comprising the following steps: Get point cloud data; encoding geometric information including positions of points of the point cloud data; generating one or more levels of detail (LODs) based on the geometric information, and selecting one or more neighboring points of each point to be attribute encoded based on the one or more LODs, wherein the selected one or more neighboring points of each point are within a maximum neighbor distance; encoding attribute information of each point based on one or more neighboring points of each selected point, wherein the maximum neighboring point distance is obtained by multiplying a first basic neighboring point distance by a maximum neighboring point range, wherein the maximum neighboring point range is calculated according to a density value of the point cloud data, wherein the density value of the point cloud data is estimated by dividing a second basic neighboring point distance for attribute encoding by a diagonal length of a bounding box of the point cloud data, and wherein the maximum neighboring point range is the number of neighboring nodes around each point; and Sending coded geometry information, coded attribute information, and signaling information, The signaling information includes information related to the maximum neighboring point range.

2. The point cloud data sending method according to claim 1, in, When the one or more LODs are generated based on an octree, the first basic neighbor distance is determined based on a diagonal distance of a node at a specific LOD.

3. A point cloud data sending device, the point cloud data sending device comprising: An acquisition unit, the acquisition unit is configured to acquire point cloud data; a geometry encoder configured to encode geometric information including positions of points of the point cloud data; an attribute encoder configured to generate one or more levels of detail (LODs) based on the geometric information, select one or more neighboring points of each point to be attribute encoded based on the one or more LODs, and encode attribute information of each point based on the selected one or more neighboring points of each point, wherein the selected one or more neighboring points of each point are located within a maximum neighboring point distance, wherein the maximum neighboring point distance is obtained by multiplying a first basic neighboring point distance by a maximum neighboring point range, wherein the maximum neighboring point range is calculated based on a density value of the point cloud data, wherein the density value of the point cloud data is estimated by dividing a second basic neighboring point distance for attribute encoding by a diagonal length of a bounding box of the point cloud data, and wherein the maximum neighboring point range is the number of neighboring nodes around each point; and A transmitter is configured to transmit encoded geometry information, encoded attribute information, and signaling information, wherein the signaling information includes information related to the maximum neighboring point range.

4. The point cloud data transmitting device according to claim 3, in, When the one or more LODs are generated based on an octree, the first basic neighbor distance is determined based on a diagonal distance of a node at a specific LOD.

5. A point cloud data receiving device, the point cloud data receiving device comprising: a receiver configured to receive geometric information, attribute information, and signaling information; a geometry decoder configured to decode the geometry information based on the signaling information; an attribute decoder configured to decode attribute information of each point based on one or more neighboring points of each point and the signaling information, wherein one or more neighboring points of each point to be attribute decoded are selected based on one or more levels of detail (LODs), wherein the one or more LODs are generated based on the geometric information, wherein the selected one or more neighboring points of each point are located within a maximum neighboring point distance, wherein the maximum neighboring point distance is obtained by multiplying a first basic neighboring point distance by a maximum neighboring point range, wherein the maximum neighboring point range is calculated based on a density value of the point cloud data, wherein the density value of the point cloud data is estimated by dividing a second basic neighboring point distance for attribute decoding by a diagonal length of a bounding box of the point cloud data, wherein the maximum neighboring point range is the number of neighboring nodes around each point, and wherein the signaling information includes information related to the maximum neighboring point range; and A renderer is configured to render point cloud data restored based on the decoded geometric information and the decoded attribute information.

6. The point cloud data receiving device according to claim 5, in, When the one or more LODs are generated based on an octree, the first basic neighbor distance is determined based on a diagonal distance of a node at a specific LOD.

7. A method for receiving point cloud data, the method comprising the following steps: receiving geometric information, attribute information and signaling information; Decoding the geometric information based on the signaling information; Decoding attribute information of each point based on one or more neighboring points of each point and the signaling information, wherein one or more neighboring points of each point to be attribute-decoded are selected based on one or more levels of detail (LODs), wherein the one or more LODs are generated based on the geometric information, wherein the selected one or more neighboring points of each point are located within a maximum neighboring point distance, wherein the maximum neighboring point distance is obtained by multiplying a first basic neighboring point distance by a maximum neighboring point range, wherein the maximum neighboring point range is calculated based on a density value of the point cloud data, wherein the density value of the point cloud data is estimated by dividing a second basic neighboring point distance for attribute decoding by a diagonal length of a bounding box of the point cloud data, wherein the maximum neighboring point range is the number of neighboring nodes around each point, and wherein the signaling information includes information related to the maximum neighboring point range; and Rendering is performed based on the point cloud data restored based on the decoded geometry information and the decoded attribute information.

8. The point cloud data receiving method according to claim 7, in, When the one or more LODs are generated based on an octree, the first basic neighbor distance is determined based on a diagonal distance of a node at a specific LOD.

Images