Point cloud data encoding device, point cloud data encoding method, point cloud data decoding device, and point cloud data decoding method

Abstract

A decoding method according to embodiments may comprise the steps of: decoding geometry data of point cloud data in a bitstream; and decoding attribute data of the point cloud data. An encoding method according to embodiments may comprise the steps of: encoding geometry data of point cloud data; and encoding attribute data of the point cloud data.

Description

Point cloud data encoding device, point cloud data encoding method, point cloud data decoding device and point cloud data decoding method

[0001] The embodiments relate to a method and apparatus for processing point cloud content.

[0002] Point cloud content is content represented as a point cloud, which is a set of points belonging to a coordinate system that represents three-dimensional space. Point cloud content can represent three-dimensional media and is used to provide various services such as VR (Virtual Reality), AR (Augmented Reality), MR (Mixed Reality), and autonomous driving services. However, representing point cloud content requires tens of thousands to hundreds of thousands of point data points. Therefore, a method is required to efficiently process a vast amount of point data.

[0003] The embodiments provide an apparatus and a method for efficiently processing point cloud data. The embodiments provide a method and apparatus for processing point cloud data to address latency and encoding / decoding complexity.

[0004] However, the scope of rights of the embodiments is not limited to the technical problems described above, and may be extended to other technical problems that a person skilled in the art can infer based on the entire content described.

[0005] A decoding method according to the embodiments may include the step of decoding geometry data of point cloud data within a bitstream; and the step of decoding attribute data of point cloud data. An encoding method according to the embodiments may include the step of encoding geometry data of point cloud data; and the step of encoding attribute data of point cloud data.

[0006] The device and method according to the embodiments can process point cloud data with high efficiency.

[0007] The device and method according to the embodiments can provide a high-quality point cloud service.

[0008] The device and method according to the embodiments can provide point cloud content for providing general-purpose services such as VR services and autonomous driving services.

[0009] Drawings are included to further understand the embodiments, and the drawings illustrate the embodiments along with descriptions related to the embodiments. For a better understanding of the various embodiments described below, one must refer to the description of the embodiments below in relation to the following drawings, which include parts corresponding to similar reference numerals throughout the drawings.

[0010] FIG. 1 shows an example of a point cloud content provision system according to embodiments.

[0011] FIG. 2 is a block diagram illustrating a point cloud content provision operation according to embodiments.

[0012] FIG. 3 shows an example of a point cloud encoder according to embodiments.

[0013] FIG. 4 shows examples of octree and occupancy codes according to embodiments.

[0014] Figure 5 shows an example of a point configuration by LOD according to embodiments.

[0015] Figure 6 shows an example of a point configuration by LOD according to embodiments.

[0016] FIG. 7 shows an example of a point cloud decoder according to embodiments.

[0017] FIG. 8 is an example of a transmission device according to embodiments.

[0018] FIG. 9 is an example of a receiving device according to embodiments.

[0019] FIG. 10 shows an example of a structure that can be linked with a point cloud data transmission / reception method / device according to embodiments.

[0020] FIG. 11 illustrates a method for generating a colored octree with attribute duplication removed according to embodiments.

[0021] FIG. 12 illustrates a leaf node level colorization method according to embodiments.

[0022] FIG. 13 illustrates a neighbor search method according to embodiments.

[0023] FIG. 14 illustrates a method for removing duplicate attributes according to embodiments.

[0024] FIG. 15 illustrates the decoder operation according to the embodiments.

[0025] FIG. 16 illustrates the process of matching attributes to octree nodes according to the embodiments.

[0026] FIG. 17 illustrates the attribute-node matching and location estimation process according to the embodiments.

[0027] FIG. 18 shows attribute parameter set syntax according to embodiments.

[0028] FIG. 19 shows the general attribute bitstream and attribute slice header syntax according to the embodiments.

[0029] FIG. 20 shows attribute slice data syntax according to embodiments.

[0030] FIG. 21 shows the encoder configuration according to the embodiments.

[0031] FIG. 22 shows a decoder configuration according to embodiments.

[0032] FIG. 23 shows block group subsampling (minGrpPts = 3) according to the embodiments.

[0033] FIG. 24 illustrates a LoD configuration method for selecting a node at a fixed position according to embodiments.

[0034] FIG. 25 illustrates a LoD configuration method for selecting a node at a fixed position according to embodiments based on a Molton code.

[0035] FIG. 26 illustrates a LoD configuration method for alternately selecting the location of nodes according to embodiments.

[0036] FIG. 27 illustrates a LoD configuration method for alternately selecting node positions according to embodiments based on a Molton code.

[0037] FIG. 28 illustrates a encoding method according to embodiments.

[0038] FIG. 29 illustrates a decoding method according to embodiments.

[0039] Preferred embodiments of the embodiments are described in detail, and examples thereof are shown in the accompanying drawings. The following detailed description, with reference to the accompanying drawings, is intended to describe preferred embodiments of the embodiments rather than merely embodiments that may be implemented according to the embodiments. The following detailed description includes details to provide a thorough understanding of the embodiments. However, it is obvious to those skilled in the art that the embodiments may be practiced without these details.

[0040] Most terms used in the embodiments are selected from those commonly used in the field, but some terms are chosen at the applicant's discretion, and their meanings are described in detail in the following description as necessary. Accordingly, the embodiments should be understood based on the intended meaning of the terms, rather than their mere names or meanings.

[0041] FIG. 1 shows an example of a point cloud content provision system according to embodiments.

[0042] The point cloud content providing system illustrated in FIG. 1 may include a transmission device (10000) and a reception device (10004). The transmission device (10000) and the reception device (10004) can communicate via wired or wireless means to transmit and receive point cloud data.

[0043] A transmission device (10000) according to embodiments can acquire, process, and transmit point cloud video (or point cloud content). According to embodiments, the transmission device (10000) may include a fixed station, a base transceiver system (BTS), a network, an AI (Artificial Intelligence) device and / or system, a robot, an AR / VR / XR device and / or server, etc. Additionally, according to embodiments, the transmission device (10000) may include a device that communicates with a base station and / or other wireless devices using wireless access technology (e.g., 5G NR (New RAT), LTE (Long Term Evolution)), a robot, a vehicle, an AR / VR / XR device, a mobile device, a home appliance, an IoT (Internet of Things) device, an AI device / server, etc.

[0044] A transmission device (10000) according to embodiments includes a point cloud video acquisition unit (10001), a point cloud video encoder (10002), and / or a transmitter (or communication module), 10003.

[0045] A point cloud video acquisition unit (10001) according to the embodiments acquires a point cloud video through processing steps such as capture, synthesis, or generation. The point cloud video is a point cloud content represented as a point cloud, which is a set of points located in a three-dimensional space, and may be referred to as point cloud video data, etc. The point cloud video according to the embodiments may include one or more frames. A frame represents a still image / picture. Accordingly, the point cloud video may include a point cloud image / frame / picture and may be referred to as any one of a point cloud image, a frame, and a picture.

[0046] A point cloud video encoder (10002) according to the embodiments encodes the obtained point cloud video data. The point cloud video encoder (10002) can encode the point cloud video data based on point cloud compression coding. The point cloud compression coding according to the embodiments may include Geometry-based Point Cloud Compression (G-PCC) coding and / or Video-based Point Cloud Compression (V-PCC) coding or next-generation coding. Furthermore, the point cloud compression coding according to the embodiments is not limited to the embodiments described above. The point cloud video encoder (10002) can output a bitstream containing the encoded point cloud video data. The bitstream may include not only the encoded point cloud video data but also signaling information related to the encoding of the point cloud video data.

[0047] A transmitter (10003) according to the embodiments transmits a bitstream containing encoded point cloud video data. The bitstream according to the embodiments is encapsulated into 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 illustrated in the drawings, the transmission device (10000) may include an encapsulation unit (or encapsulation module) that performs an encapsulation operation. Additionally, according to the embodiments, the encapsulation unit may be included in the transmitter (10003). According to the embodiments, the file or segment may be transmitted to a receiving device (10004) via a network or stored on a digital storage medium (e.g., USB, SD, CD, DVD, Blu-ray, HDD, SSD, etc.). The transmitter (10003) according to the embodiments can communicate wired or wirelessly with the receiving device (10004) (or receiver (10005)) via a network such as 4G, 5G, or 6G. Additionally, the transmitter (10003) can perform necessary data processing operations according to a network system (e.g., a communication network system such as 4G, 5G, 6G, etc.). Additionally, the transmission device (10000) can transmit encapsulated data according to an on-demand method.

[0048] A receiving device (10004) according to embodiments includes a receiver (10005), a point cloud video decoder (10006), and / or a renderer (10007). According to embodiments, the receiving device (10004) may include a device, robot, vehicle, AR / VR / XR device, mobile device, home appliance, IoT (Internet of Thing) device, AI device / server, etc., that communicates with a base station and / or other wireless device using wireless access technology (e.g., 5G NR (New RAT), LTE (Long Term Evolution)).

[0049] A receiver (10005) according to the embodiments receives a bitstream containing point cloud video data or a file / segment containing the bitstream from a network or a storage medium. The receiver (10005) can perform necessary data processing operations according to a network system (e.g., a communication network system such as 4G, 5G, 6G, etc.). The receiver (10005) according to the embodiments can output a bitstream by decapsulating the received file / segment. Additionally, according to the embodiments, the receiver (10005) may include a decapsulation unit (or decapsulation module) for performing a decapsulation operation. Additionally, the decapsulation unit may be implemented as an element (or component) separate from the receiver (10005).

[0050] A point cloud video decoder (10006) decodes a bitstream containing point cloud video data. The point cloud video decoder (10006) can decode the point cloud video data according to the way the point cloud video data is encoded (e.g., the reverse process of the operation of a point cloud video encoder (10002)). Accordingly, the point cloud video decoder (10006) can decode the point cloud video data by performing point cloud decompression coding, which is the reverse process of point cloud compression. Point cloud decompression coding includes G-PCC coding.

[0051] The renderer (10007) renders the decoded point cloud video data. The renderer (10007) can render not only the point cloud video data but also audio data to output point cloud content. According to embodiments, the renderer (10007) may include a display for displaying the point cloud content. According to embodiments, the display may not be included in the renderer (10007) but may be implemented as a separate device or component.

[0052] The arrows indicated by dotted lines in the drawing represent the transmission path of feedback information obtained from the receiving device (10004). The feedback information is information intended to reflect interaction with a user consuming point cloud content, and includes user information (e.g., head orientation information), viewport information, etc. In particular, if the point cloud content is content for a service requiring interaction with a user (e.g., autonomous driving service, etc.), the feedback information may be transmitted to the content transmitting side (e.g., the transmitting device (10000)) and / or the service provider. Depending on the embodiments, the feedback information may be used in the receiving device (10004) as well as the transmitting device (10000), or it may not be provided.

[0053] Head orientation information according to the embodiments is information regarding the user's head position, direction, angle, movement, etc. The receiving device (10004) according to the embodiments can calculate viewport information based on the head orientation information. Viewport information is information about the area of ​​the point cloud video that the user is looking at. The viewpoint refers to the point where the user is looking at the point cloud video, and may mean the exact 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, the receiving device (10004) can extract viewport information based on the vertical or horizontal FOV supported by the device in addition to the head orientation information. In addition, the receiving device (10004) performs gaze analysis, etc., to check the user's point cloud consumption method, the point cloud video area the user is looking at, the gaze time, etc. According to embodiments, the receiving device (10004) may transmit feedback information including gaze analysis results to the transmitting device (10000). According to embodiments, the feedback information may be obtained during the rendering and / or display process. According to embodiments, the feedback information may be obtained by one or more sensors included in the receiving device (10004). Also, according to embodiments, the feedback information may be obtained by the renderer (10007) or a separate external element (or device, component, etc.). The dotted line in FIG. 1 indicates the process of transmitting the feedback information obtained from the renderer (10007). The point cloud content providing system may process (encode / decode) point cloud data based on the feedback information. Accordingly, the point cloud video data decoder (10006) may perform a decoding operation based on the feedback information.Additionally, the receiving device (10004) can transmit feedback information to the transmitting device (10000). The transmitting device (10000) (or the point cloud video data encoder (10002)) can perform an encoding operation based on the feedback information. Thus, the point cloud content providing system can efficiently process necessary data (e.g., point cloud data corresponding to the user's head position) based on the feedback information without processing (encoding / decoding) all point cloud data, and provide point cloud content to the user.

[0054] According to embodiments, the transmission device (10000) may be referred to as an encoder, transmission device, transmitter, etc., and the receiving device (10004) may be referred to as a decoder, receiving device, receiver, etc.

[0055] Point cloud data processed in the point cloud content providing system of FIG. 1 according to embodiments (processed through a series of processes of acquisition / encoding / transmission / decoding / rendering) may be referred to as point cloud content data or point cloud video data. According to embodiments, point cloud content data may be used as a concept including metadata or signaling information related to point cloud data.

[0056] The elements of the point cloud content delivery system illustrated in FIG. 1 can be implemented using hardware, software, processors, and / or combinations thereof.

[0057] FIG. 2 is a block diagram illustrating a point cloud content provision operation according to embodiments.

[0058] The block diagram of FIG. 2 illustrates the operation of the point cloud content provision system described in FIG. 1. As described above, the point cloud content provision system can process point cloud data based on point cloud compression coding (e.g., G-PCC).

[0059] A point cloud content providing system according to the embodiments (e.g., a point cloud transmission device (10000) or a point cloud video acquisition unit (10001)) can acquire a point cloud video (20000). The point cloud video is represented as a point cloud belonging to a coordinate system representing a three-dimensional space. The point cloud video according to the embodiments may include a Ply (Polygon File format or the Stanford Triangle format) file. If 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 the geometry and / or attributes of the points. The geometry includes the positions of the points. The position of each point may be represented by parameters (e.g., values ​​of the X-axis, Y-axis, and Z-axis, respectively) representing a three-dimensional coordinate system (e.g., a coordinate system consisting of XYZ axes). Attributes include attributes of points (e.g., texture information, color (YCbCr or RGB), reflectance (r), transparency, etc. of each point). A point has one or more attributes (or properties). For example, a point may have one attribute which is color, or two attributes which are color and reflectance. According to embodiments, geometry may be referred to as positions, geometry information, geometry data, etc., and attributes may be referred to as attributes, attribute information, attribute data, etc.In addition, a point cloud content provision system (e.g., a point cloud transmission device (10000) or a point cloud video acquisition unit (10001)) can obtain point cloud data from information related to the acquisition process of point cloud video (e.g., depth information, color information, etc.).

[0060] A point cloud content providing system according to embodiments (e.g., a transmission device (10000) or a point cloud video encoder (10002)) can encode point cloud data (20001). The point cloud content providing system can encode point cloud data based on point cloud compression coding. As described above, point cloud data may include geometry and attributes of points. Accordingly, the point cloud content providing system can output a geometry bitstream by performing geometry encoding to encode geometry. The point cloud content providing system can output an attribute bitstream by performing attribute encoding to encode attributes. According to embodiments, the point cloud content providing system can perform attribute encoding based on geometry encoding. The geometry bitstream and attribute bitstream according to embodiments can be multiplexed and output as a single bitstream. The bitstream according to the embodiments may further include signaling information related to geometry encoding and attribute encoding.

[0061] A point cloud content providing system according to embodiments (e.g., a transmission device (10000) or a transmitter (10003)) can transmit encoded point cloud data (20002). As described in FIG. 1, the encoded point cloud data can be represented as a geometry bitstream and an attribute bitstream. Additionally, the encoded point cloud data can be transmitted in the form of a bitstream along with signaling information related to the encoding of the point cloud data (e.g., signaling information related to geometry encoding and attribute encoding). Additionally, the point cloud content providing system can encapsulate the bitstream transmitting the encoded point cloud data and transmit it in the form of a file or segment.

[0062] A point cloud content providing system according to embodiments (e.g., a receiving device (10004) or a receiver (10005)) can receive a bitstream containing encoded point cloud data. Additionally, the point cloud content providing system (e.g., a receiving device (10004) or a receiver (10005)) can demultiplex the bitstream.

[0063] A point cloud content providing system (e.g., a receiving device (10004) or a point cloud video decoder (10005)) can decode encoded point cloud data (e.g., a geometry bitstream, an attribute bitstream) transmitted as a bitstream. A point cloud content providing system (e.g., a receiving device (10004) or a point cloud video decoder (10005)) can decode point cloud video data based on signaling information related to the encoding of point cloud video data included in the bitstream. A point cloud content providing system (e.g., a receiving device (10004) or a point cloud video decoder (10005)) can decode the geometry bitstream to restore the positions (geometry) of the points. A point cloud content providing system can decode the attribute bitstream based on the restored geometry to restore the attributes of the points. A point cloud content delivery system (e.g., a receiving device (10004) or a point cloud video decoder (10005)) can restore a point cloud video based on positions according to the restored geometry and decoded attributes.

[0064] A point cloud content providing system according to embodiments (e.g., a receiving device (10004) or a renderer (10007)) can render decoded point cloud data (20004). The point cloud content providing system (e.g., a receiving device (10004) or a renderer (10007)) can render geometry and attributes decoded through a decoding process according to various rendering methods. Points of the point cloud content may be rendered as vertices having a certain thickness, cubes having a specific minimum size with the vertex location as the center, or circles with the vertex location as the center, etc. All or part of the rendered point cloud content is provided to a user through a display (e.g., a VR / AR display, a general display, etc.).

[0065] A point cloud content providing system (e.g., a receiving device (10004)) according to the embodiments can obtain feedback information (20005). The point cloud content providing system can encode and / or decode point cloud data based on the feedback information. Since the feedback information and the operation of the point cloud content providing system according to the embodiments are the same as the feedback information and operation described in FIG. 1, a detailed description is omitted.

[0066] FIG. 3 shows an example of a point cloud encoder according to embodiments.

[0067] FIG. 3 shows an example of the point cloud video encoder (10002) of FIG. 1. The point cloud encoder reconstructs point cloud data (e.g., positions and / or attributes of points) and performs encoding operations to adjust the quality of point cloud content (e.g., lossless, lossy, near-lossless) according to network conditions or applications. If the total size of the point cloud content is large (e.g., point cloud content of 60 Gbps in the case of 30 fps), the point cloud content delivery system may not be able to stream the content in real time. Therefore, the point cloud content delivery system may reconstruct the point cloud content based on a maximum target bitrate to provide it according to the network environment.

[0068] As described in FIGS. 1 and 2, the point cloud encoder can perform geometry encoding and attribute encoding. Geometry encoding is performed before attribute encoding.

[0069] The point cloud encoder according to the embodiments comprises a coordinate system transformation unit (Transformation Coordinates, 30000), a quantization unit (Quantize and Remove Points (Voxelize), 30001), an octree analysis unit (Analyze Octree, 30002), a surface approximation analysis unit (Analyze Surface Approximation, 30003), an arithmetic encoder (Arithmetic Encode, 30004), a geometry reconstruction unit (Reconstruct Geometry, 30005), a color transformation unit (Transform Colors, 30006), an attribute transformation unit (Transfer Attributes, 30007), a RAHT transformation unit (30008), an LOD generation unit (Generated LOD, 30009), a lifting transformation unit (Lifting) (30010), and a coefficient quantization unit (Quantize Coefficients, 30011). Includes an and / or arithmetic encoder (Arithmetic Encode, 30012).

[0070] The coordinate system transformation unit (30000), quantization unit (30001), octree analysis unit (30002), surface approximation analysis unit (30003), arismetic encoder (30004), and geometry reconstruction unit (30005) can perform geometry encoding. Geometry encoding according to the embodiments may include octree geometry coding, direct coding, trisoup geometry encoding, and entropy encoding. Direct coding and trisoup geometry encoding are applied optionally or in combination. Additionally, geometry encoding is not limited to the above examples.

[0071] As illustrated in the drawings, the coordinate system conversion unit (30000) according to the embodiments receives positions and converts them into a coordinate system. For example, the positions can be converted into position information in a three-dimensional space (e.g., a three-dimensional space expressed in an XYZ coordinate system). The position information in the three-dimensional space according to the embodiments may be referred to as geometry information.

[0072] The quantization unit (30001) according to the embodiments quantizes the geometry. For example, the quantization unit (30001) can quantize points based on the minimum position values ​​of all points (e.g., minimum values ​​on each axis for the X-axis, Y-axis, and Z-axis). The quantization unit (30001) performs a quantization operation to find the nearest integer value by multiplying the difference between the minimum position value and the position value of each point by a preset quantization scale value and then performing rounding down or rounding up. Thus, one or more points may have the same quantized position (or position value). The quantization unit (30001) according to the embodiments performs voxelization based on the quantized positions to reconstruct the quantized points. Just as the minimum unit containing 2D image / video information is a pixel, the points of the point cloud content (or 3D point cloud video) according to the embodiments may be contained in one or more voxels. A voxel is a combination of volume and pixel, and refers to a three-dimensional cubic space that is generated when a three-dimensional space is divided into units (unit=1.0) based on axes representing the three-dimensional space (e.g., X-axis, Y-axis, Z-axis). The quantization unit (40001) can match groups of points in the three-dimensional space to voxels. According to embodiments, a single voxel may contain only one point. According to embodiments, a single voxel may contain one or more points. In addition, to represent a single voxel as a single point, the position of the center of the voxel can be set based on the positions of one or more points included in the voxel. In this case, the attributes of all positions included in the voxel can be combined and assigned to the voxel.

[0073] The octree analysis unit (30002) according to the embodiments performs octree geometry coding (or octree coding) to represent the voxels in an octree structure. The octree structure represents points matched to the voxels based on an octree structure.

[0074] The surface approximation analysis unit (30003) according to the embodiments can analyze and approximate an octree. The octree analysis and approximation according to the embodiments is a process of analyzing to voxelize an area containing multiple points in order to efficiently provide octree and voxelization.

[0075] An arithmetic encoder (30004) according to the embodiments entropy-encodes an octree and / or an approximated octree. For example, the encoding method includes an arithmetic encoding method. As a result of the encoding, a geometry bitstream is generated.

[0076] The color conversion unit (30006), attribute conversion unit (30007), RAHT conversion unit (30008), LOD generation unit (30009), lifting conversion unit (30010), coefficient quantization unit (30011) and / or arismetic encoder (30012) perform attribute encoding. As described above, a point may have one or more attributes. The attribute encoding according to the embodiments is applied equally to the attributes of a point. However, if a single attribute (e.g., color) includes one or more elements, independent attribute encoding is applied to each element. The attribute encoding according to the embodiments may include color conversion coding, attribute conversion coding, Region Adaptive Hierarchial Transform (RAHT) coding, prediction transformation (Interpolaration-based hierarchical nearest-neighbour prediction-Prediction Transform) coding, and lifting transformation (interpolation-based hierarchical nearest-neighbour prediction with an update / lifting step (Lifting Transform)) coding. Depending on the point cloud content, the above-described RAHT coding, prediction transformation coding, and lifting transformation coding may be used optionally, or a combination of one or more of the codings may be used. Furthermore, the attribute encoding according to the embodiments is not limited to the examples described above.

[0077] The color conversion unit (30006) according to the embodiments performs color conversion coding that converts color values ​​(or textures) included in attributes. For example, the color conversion unit (30006) can convert the format of color information (e.g., convert from RGB to YCbCr). The operation of the color conversion unit (30006) according to the embodiments may be applied optionally depending on the color values ​​included in attributes.

[0078] The geometry reconstruction unit (30005) according to the embodiments reconstructs (decompresses) an octree and / or an approximated octree. The geometry reconstruction unit (30005) reconstructs an octree / voxel based on the results of analyzing the distribution of points. The reconstructed octree / voxel may be referred to as the reconstructed geometry (or restored geometry).

[0079] The attribute transformation unit (30007) according to the embodiments performs attribute transformation that transforms attributes based on positions where geometry encoding has not been performed and / or reconstructed geometry. As described above, since attributes are dependent on geometry, the attribute transformation unit (30007) can transform attributes based on reconstructed geometry information. For example, the attribute transformation unit (30007) can transform the attributes of a point at a position based on the position value of a point included in a voxel. As described above, when the position of the center point of a voxel is set based on the positions of one or more points included in a voxel, the attribute transformation unit (30007) transforms the attributes of one or more points. When trisoop geometry encoding is performed, the attribute conversion unit (30007) can convert attributes based on the trisoop geometry encoding.

[0080] The attribute transformation unit (30007) can perform attribute transformation by calculating the average value of attributes or attribute values ​​(e.g., the color or reflectance of each point) of neighboring points within a specific location / radius from the position (or position value) of the center point of each voxel. The attribute transformation unit (30007) can apply a weight based on the distance from the center point to each point when calculating the average value. Thus, each voxel has a position and a calculated attribute (or attribute value).

[0081] The attribute conversion unit (30007) can search for neighboring points within a specific location / radius from the position of the center point of each voxel based on a KD tree or a Molton code. A KD tree is a binary search tree that supports a data structure capable of managing points based on their positions to enable rapid Nearest Neighbor Search (NNS). A Molton code is generated by representing the coordinate values ​​(e.g., (x, y, z)) representing the 3D positions of all points as bit values ​​and mixing the bits. For example, if the coordinate values ​​representing the position of a point are (5, 9, 1), the bit values ​​of the coordinate values ​​are (0101, 1001, 0001). When the bit values ​​are mixed according to the bit indices in the order of z, y, and x, it becomes 010001000111. When this value is represented in decimal, it becomes 1095. That is, the Molton code value of the point with coordinates (5, 9, 1) is 1095. The attribute transformation unit (30007) sorts the points based on the Molton code value and can perform shortest neighbor search (NNS) through a depth-first traversal process. After the attribute transformation operation, if shortest neighbor search (NNS) is required in other transformation processes for attribute coding, a KD tree or Molton code is utilized.

[0082] As shown in the drawing, the converted attributes are input to the RAHT conversion unit (30008) and / or LOD generation unit (30009).

[0083] The RAHT transformation unit (30008) according to the embodiments performs RAHT coding to predict attribute information based on reconstructed geometry information. For example, the RAHT transformation unit (30008) can predict attribute information of a node at an upper level of the octree based on attribute information associated with a node at a lower level of the octree.

[0084] The LOD generation unit (30009) according to the embodiments generates a Level of Detail (LOD) to perform predictive transformation coding. The LOD according to the embodiments represents the degree of detail of the point cloud content, and indicates that the smaller the LOD value, the lower the detail of the point cloud content, and the larger the LOD value, the higher the detail of the point cloud content. Points can be classified according to the LOD.

[0085] The lifting transformation unit (30010) according to the embodiments performs lifting transformation coding that transforms the attributes of the point cloud based on weights. As described above, the lifting transformation coding may be applied optionally.

[0086] The coefficient quantization unit (30011) according to the embodiments quantizes attribute-coded attributes based on coefficients.

[0087] An arismetic encoder (30012) according to the embodiments encodes quantized attributes based on arismetic coding.

[0088] The elements of the point cloud encoder of FIG. 3 may be implemented in hardware, software, firmware, or a combination thereof, comprising one or more processors or integrated circuits configured to communicate with one or more memories included in the point cloud providing device, although not illustrated in the drawing. One or more processors may perform at least one of the operations and / or functions of the elements of the point cloud encoder of FIG. 3 described above. Additionally, one or more processors may operate or execute a set of software programs and / or instructions for performing the operations and / or functions of the elements of the point cloud encoder of FIG. 3. One or more memories according to the embodiments may include high-speed random access memory and may include non-volatile memory (e.g., one or more magnetic disk storage devices, flash memory devices, or other non-volatile solid-state memory devices).

[0089] FIG. 4 shows examples of octree and occupancy codes according to embodiments.

[0090] As described in FIGS. 1 to 3, a point cloud content providing system (point cloud video encoder (10002)) or a point cloud encoder (e.g., an octree analysis unit (30002)) performs octree geometry coding based on an octree structure (or octree coding) to efficiently manage the area and / or position of a voxel.

[0091] The top of FIG. 4 shows an octree structure. The three-dimensional space of the point cloud content according to the embodiments is represented by the axes of the coordinate system (e.g., X-axis, Y-axis, Z-axis). The octree structure has two poles (0,0,0) and (2 d , 2 d , 2 d It is generated by recursively subdividing the bounding box (cubical axis-aligned bounding box) defined by ). 2d can be set to the value that constitutes the smallest bounding box enclosing all points of the point cloud content (or point cloud video). d represents the depth of the octree. The value of d is determined according to the following equation. In the equation below, (x int n , y int n , z int n ) represents the positions (or position values) of quantized points.

[0092] d=Ceil(Log2(Max(x int n ,y int n ,z int n ,n=1,… ,N)+1))

[0093] As illustrated in the middle of the top of Fig. 4, the entire three-dimensional space can be divided into eight spaces according to the division. Each divided space is represented as a cube having six faces. As illustrated in the right of the top of Fig. 4, each of the eight spaces is further divided based on the axes of the coordinate system (e.g., X-axis, Y-axis, Z-axis). Thus, each space is again divided into eight smaller spaces. The divided smaller spaces are also represented as cubes having six faces. This division method is applied until the leaf nodes of the octree become voxels.

[0094] The bottom of Fig. 4 shows the occupancy code of an octree. The occupancy code of an octree is generated to indicate whether each of the eight partitioned spaces resulting from the partitioning of a single space contains at least one point. Therefore, one occupancy code is represented by eight child nodes. Each child node represents the occupancy of the partitioned space, and the child node has a value of 1 bit. Thus, the occupancy code is represented as an 8-bit code. That is, if the space corresponding to the child node contains at least one point, the node has a value of 1. If the space corresponding to the child node does not contain a point (empty), the node has a value of 0. Since the occupancy code shown in Fig. 4 is 00100001, it indicates that the spaces corresponding to the 3rd and 8th child nodes among the eight child nodes each contain at least one point. As illustrated in the drawing, the 3rd child node and the 8th child node each have 8 child nodes, and each child node is represented by an 8-bit Occupancy code. The drawing indicates that the Occupancy code of the 3rd child node is 10000111 and the Occupancy code of the 8th child node is 01001111. A point cloud encoder according to the embodiments (e.g., an arismetic encoder (30004)) can entropy-encode the Occupancy code. Additionally, to increase compression efficiency, the point cloud encoder can intra- / inter-encode the Occupancy code. A receiving device according to the embodiments (e.g., a receiving device (10004) or a point cloud video decoder (10006)) reconstructs the octree based on the Occupancy code.

[0095] A point cloud encoder according to the embodiments (e.g., the point cloud encoder of FIG. 3, or the octree analysis unit (30002)) can perform voxelization and octree coding to store the positions of the points. However, since points in a three-dimensional space are not always evenly distributed, there may be specific areas where few points exist. Therefore, performing voxelization on the entire three-dimensional space is inefficient. For example, if there are almost no points in a specific area, there is no need to perform voxelization up to that area.

[0096] Accordingly, the point cloud encoder according to the embodiments can perform direct coding, which directly codes the positions of points included in a specific region (or nodes excluding leaf nodes of an octree) without performing voxelization on the aforementioned specific region. The coordinates of the points directly coded according to the embodiments are referred to as the Direct Coding Mode (DCM). Additionally, the point cloud encoder according to the embodiments can perform trisoup geometry encoding, which reconstructs the positions of points within a specific region (or node) based on voxels using a surface model. Trisoup geometry encoding is a geometry encoding that represents an object as a series of triangle meshes. Therefore, the point cloud decoder can generate a point cloud from the mesh surface. Direct coding and trisoup geometry encoding according to the embodiments may be performed optionally. In addition, direct coding and trisoop geometry encoding according to the embodiments can be performed in combination with octree geometry coding (or octree coding).

[0097] To perform direct coding, the option to use direct mode for applying direct coding must be enabled, the node to which direct coding is to be applied must not be a leaf node, and there must be points within a specific node that are below a threshold. In addition, the total number of points subject to direct coding must not exceed a preset threshold. If the above conditions are satisfied, the point cloud encoder (or arismetic encoder (30004)) according to the embodiments can entropy-code the positions (or position values) of the points.

[0098] A point cloud encoder according to the embodiments (e.g., a surface approximation analysis unit (30003)) can determine a specific level of an octree (where the level is smaller than the depth d of the octree) and, starting from that level, perform trisoop geometry encoding to reconstruct the position of points within a node region based on voxels using a surface model (trisoop mode). The point cloud encoder according to the embodiments can specify the level to which trisoop geometry encoding is applied. For example, if the specified level is equal to the depth of the octree, the point cloud encoder does not operate in trisoop mode. That is, the point cloud encoder according to the embodiments can operate in trisoop mode only when the specified level is smaller than the depth value of the octree. A three-dimensional cubic region of nodes at a specified level according to the embodiments is referred to as a block. A block may include one or more voxels. A block or a voxel may correspond to a brick. Within each block, geometry is represented as a surface. A surface according to the embodiments may intersect each edge of the block at most once.

[0099] Since one block has 12 edges, there are at least 12 intersection points within one block. Each intersection point is referred to as a vertex. A vertex along an edge is detected if there is at least one occupied voxel adjacent to that edge among all blocks sharing that edge. An occupied voxel according to the embodiments means a voxel containing a point. The position of a vertex detected along an edge is the average position along the edge of all voxels adjacent to that edge among all blocks sharing that edge.

[0100] When a vertex is detected, the point cloud encoder according to the embodiments can entropy-code the edge start point (x, y, z), edge direction vector (Δx, Δy, Δz), and vertex position value (relative position value within the edge). When trisoop geometry encoding is applied, the point cloud encoder according to the embodiments (e.g., geometry reconstruction unit (30005)) can generate restored geometry (reconstructed geometry) by performing triangle reconstruction, up-sampling, and voxelization processes.

[0101] The vertices located on the edges of the block determine the surface passing through the block. The surface according to the embodiments is a non-planar polygon. The triangle reconstruction process reconstructs the surface represented by triangles based on the edge start point, the edge direction vector, and the vertex position value. The triangle reconstruction process is as follows: ① calculate the centroid value of each vertex, ② subtract the centroid value from each vertex value, ③ square the result, and add all the result together.

[0102]

[0103] The minimum sum is calculated, and a projection process is performed along the axis where the minimum value is located. For example, if the x-element is at its minimum, each vertex is projected along the x-axis relative to the center of the block and onto the (y, z) plane. If the resulting value from projection onto the (y, z) plane is (ai, bi), the θ value is calculated using atan2(bi, ai), and the vertices are aligned based on this θ value. The table below shows the combinations of vertices to generate triangles based on the number of vertices. The vertices are aligned in order from 1 to n. The table below indicates that for four vertices, two triangles can be formed depending on the combination of vertices. The first triangle is composed of the 1st, 2nd, and 3rd vertices among the aligned vertices, and the second triangle can be composed of the 3rd, 4th, and 1st vertices among the aligned vertices.

[0104] Table 2-1. Triangles formed from vertices ordered 1,… ,n

[0105] n triangles

[0106] 3 (1,2,3)

[0107] 4 (1,2,3), (3,4,1)

[0108] 5 (1,2,3), (3,4,5), (5,1,3)

[0109] 6 (1,2,3), (3,4,5), (5,6,1), (1,3,5)

[0110] 7 (1,2,3), (3,4,5), (5,6,7), (7,1,3), (3,5,7)

[0111] 8 (1,2,3), (3,4,5), (5,6,7), (7,8,1), (1,3,5), (5,7,1)

[0112] 9 (1,2,3), (3,4,5), (5,6,7), (7,8,9), (9,1,3), (3,5,7), (7,9,3)

[0113] 10 (1,2,3), (3,4,5), (5,6,7), (7,8,9), (9,10,1), (1,3,5), (5,7,9), (9,1,5)

[0114] 11 (1,2,3), (3,4,5), (5,6,7), (7,8,9), (9,10,11), (11,1,3), (3,5,7), (7,9,11), (11,3,7)

[0115] 12 (1,2,3), (3,4,5), (5,6,7), (7,8,9), (9,10,11), (11,12,1), (1,3,5), (5,7,9), (9,11,1), (1,5,9)

[0116] The upsampling process is performed to voxelize by adding intermediate points along the edges of the triangle. Additional points are generated based on the upsampling factor value and the width of the block. The additional points are referred to as refined vertices. A point cloud encoder according to the embodiments can voxelize the refined vertices. Additionally, the point cloud encoder can perform attribute encoding based on the voxelized positions (or position values).

[0117] Figure 5 shows an example of a point configuration by LOD according to embodiments.

[0118] As described in FIGS. 1 to 4, the encoded geometry is reconstructed (decompressed) before attribute encoding is performed. When direct coding is applied, the geometry reconstruction operation may include changing the arrangement of the direct-coded points (e.g., placing the direct-coded points at the front of the point cloud data). When trisoop geometry encoding is applied, the geometry reconstruction process involves triangle reconstruction, upsampling, and voxelization. Since attributes depend on geometry, attribute encoding is performed based on the reconstructed geometry.

[0119] A point cloud encoder (e.g., an LOD generation unit (30009)) can reorganize points by LOD. The drawing shows point cloud content corresponding to the LOD. The left side of the drawing shows the original point cloud content. The second figure from the left in the drawing shows the distribution of points of the lowest LOD, and the rightmost figure in the drawing shows the distribution of points of the highest LOD. That is, the points of the lowest LOD are sparsely distributed, while the points of the highest LOD are densely distributed. In other words, according to the direction of the arrow at the bottom of the drawing, as the LOD increases, the spacing (or distance) between points becomes shorter.

[0120] Figure 6 shows an example of a point configuration by LOD according to embodiments.

[0121] As described in FIGS. 1 to 5, a point cloud content providing system or a point cloud encoder (e.g., a point cloud video encoder (10002), the point cloud encoder of FIG. 3, or an LOD generation unit (30009)) can generate an LOD. The LOD is generated by reorganizing points into a set of refinement levels according to a set LOD distance value (or a set of Euclidean distances). The LOD generation process is performed in a point cloud decoder as well as a point cloud encoder.

[0122] The top of Fig. 6 shows examples of points (P0 to P9) of point cloud content distributed in three-dimensional space. The Original Order in Fig. 6 represents the order of points P0 to P9 prior to LOD generation. The LOD-based Order in Fig. 6 represents the order of points following LOD generation. Points are rearranged by LOD. Additionally, higher LODs include points belonging to lower LODs. As illustrated in Fig. 6, LOD0 includes P0, P5, P4, and P2. LOD1 includes the points of LOD0 and P1, P6, and P3. LOD2 includes the points of LOD0, the points of LOD1, and P9, P8, and P7.

[0123] As described in FIG. 3, the point cloud encoder according to the embodiments can perform predictive transform coding, lifting transform coding, and RAHT transform coding selectively or in combination.

[0124] The point cloud encoder according to the embodiments can generate predictors for points and perform predictive transformation coding to set the predicted attribute (or predicted attribute value) of each point. That is, N predictors can be generated for N points. The predictor according to the embodiments can calculate a weight (=1 / distance) value based on the LOD value of each point, indexing information for neighboring points within a set distance per LOD, and the distance value to the neighboring points.

[0125] According to the embodiments, the predicted attribute (or attribute value) is set as the average value of the values ​​obtained by multiplying the attributes (or attribute values, e.g., color, reflectance, etc.) of neighboring points set in the predictor of each point by a weight (or weight value) calculated based on the distance to each neighboring point. The point cloud encoder according to the embodiments (e.g., coefficient quantization unit (30011)) can quantize and inverse quantize the residual values ​​(which may be referred to as residual attributes, residual attribute values, attribute prediction residual values, etc.) obtained by subtracting the predicted attribute (attribute value) from the attribute (attribute value) of each point. The quantization process is as shown in the following table.

[0126] Attribute prediction residuals quantization pseudo codeint 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);}}

[0127] Attribute prediction residuals inverse quantization pseudo codeint PCCInverseQuantization(int value, int quantStep) {if( quantStep ==0) {return value;} else {return value * quantStep;}}

[0128] A point cloud encoder according to the embodiments (e.g., an arismetic encoder (30012)) can entropy-code the quantized and inversely quantized residual values ​​as described above when there are neighboring points in the predictor of each point. A point cloud encoder according to the embodiments (e.g., an arismetic encoder (30012)) can entropy-code the attributes of the corresponding point without performing the process described above when there are no neighboring points in the predictor of each point.

[0129] A point cloud encoder according to the embodiments (e.g., a lifting transformation unit (30010)) can perform lifting transformation coding by generating a predictor for each point, setting the LOD calculated in the predictor, registering neighboring points, and setting weights based on the distance to neighboring points. The lifting transformation coding according to the embodiments is similar to the prediction transformation coding described above, but differs in that weights are cumulatively applied to attribute values. The process of cumulatively applying weights to attribute values ​​according to the embodiments is as follows.

[0130] 1) Create an array QW (QuantizationWight) to store the weight values ​​of each point. The initial value of all elements in QW is 1.0. Add the value obtained by multiplying the current point's predictor weight by the QW value of the predictor index of the neighboring node registered in the predictor.

[0131] 2) Lift prediction process: To calculate the predicted attribute value, the value obtained by multiplying the point's attribute value by a weight is subtracted from the existing attribute value.

[0132] 3) Create temporary arrays named updateweight and update, and initialize the temporary arrays to 0.

[0133] 4) For all predictors, the calculated weight is additionally multiplied by the weight stored in the QW corresponding to the predictor index, and the resulting weight is accumulated in the update weight array with the neighbor node index. In the update array, the value obtained by multiplying the attribute value of the neighbor node index by the calculated weight is accumulated.

[0134] 5) Lift update process: For all predictors, the attribute value of the update array is divided by the weight value of the update weight array at the predictor index, and the original attribute value is added back to the divided value.

[0135] 6) For all predictors, the predicted attribute value is calculated by additionally multiplying the attribute value updated through the lift update process by the weight (stored in QW) updated through the lift prediction process. A point cloud encoder according to the embodiments (e.g., coefficient quantizer (30011)) quantizes the predicted attribute value. Additionally, a point cloud encoder (e.g., arismetic encoder (30012)) entropies the quantized attribute value.

[0136] A point cloud encoder according to the embodiments (e.g., a RAHT transform unit (30008)) can perform RAHT transform coding to predict attributes of upper-level nodes using attributes associated with nodes at lower levels of the octree. RAHT transform coding is an example of attribute intra-coding through octree backward scanning. A point cloud encoder according to the embodiments scans from voxels to the entire region and repeats the merging process up to the root node, merging voxels into larger blocks at each step. The merging process according to the embodiments is performed only on occupied nodes. The merging process is not performed on empty nodes, and the merging process is performed on the node immediately above the empty node.

[0137] The following equation represents the RAHT transformation matrix. is level Represents the average attribute value of the voxels in. Is and It can be calculated from. and The weights of class am.

[0138]

[0139] is a low-pass value used in the merging process at the next higher level. are high-pass coefficients, and the high-pass coefficients at each step are quantized and entropy-coded (e.g., encoding of an arismetic encoder (300012)). The weights are It is calculated as. The root node is the last class It is generated as follows through.

[0140]

[0141] The gDC value is also quantized and entropy-coded, just like the high-pass coefficient.

[0142] FIG. 7 shows an example of a point cloud decoder according to embodiments.

[0143] The point cloud decoder illustrated in FIG. 7 is an example of a point cloud decoder and can perform a decoding operation, which is the reverse process of the encoding operation of the point cloud encoder described in FIG. 1 to 6.

[0144] As described in Fig. 1, the point cloud decoder can perform geometry decoding and attribute decoding. Geometry decoding is performed before attribute decoding.

[0145] A point cloud decoder according to the embodiments comprises an arithmetic decoder (7000), a synthesize octree (7001), a synthesize surface approximation (7002), a reconstruct geometry (7003), an inverse transform coordinates (7004), an arithmetic decoder (7005), an inverse quantize (7006), a RAHT transform (7007), an LOD generater (7008), an inverse lifting (7009), and / or an inverse transform colors (7010).

[0146] An arismetic decoder (7000), an octree composite unit (7001), a surface offset composite unit (7002), a geometry reconstruction unit (7003), and a coordinate system inverse transformation unit (7004) can perform geometry decoding. Geometry decoding according to the embodiments may include direct coding and trisoup geometry decoding. Direct coding and trisoup geometry decoding are applied optionally. Additionally, geometry decoding is not limited to the above examples and is performed as the reverse process of geometry encoding described in FIGS. 1 through 6.

[0147] The arismetic decoder (7000) according to the embodiments decodes the received geometry bitstream based on arismetic coding. The operation of the arismetic decoder (7000) corresponds to the reverse process of the arismetic encoder (30004).

[0148] The octree synthesis unit (7001) according to the embodiments can generate an octree by obtaining an Occupancy code from a decoded geometry bitstream (or information regarding the geometry obtained as a result of decoding). A specific description of the Occupancy code is as described in FIGS. 1 to 6.

[0149] The surface off-relation synthesis unit (7002) according to the embodiments can synthesize a surface based on the decoded geometry and / or the generated octree when trisoop geometry encoding is applied.

[0150] The geometry reconstruction unit (7003) according to the embodiments can regenerate geometry based on a surface and / or decoded geometry. As described in FIGS. 1 through 6, direct coding and trisoop geometry encoding are applied optionally. Accordingly, the geometry reconstruction unit (7003) directly retrieves and adds position information of points to which direct coding has been applied. In addition, when trisoop geometry encoding is applied, the geometry reconstruction unit (7003) can restore geometry by performing reconstruction operations of the geometry reconstruction unit (30005), such as triangle reconstruction, up-sampling, and voxelization operations. Specific details are omitted as they are the same as those described in FIG. 4. The restored geometry may include a point cloud picture or frame that does not contain attributes.

[0151] The coordinate system inverse transformation unit (7004) according to the embodiments can obtain the positions of the points by transforming the coordinate system based on the restored geometry.

[0152] The arismetic decoder (7005), inverse quantization unit (7006), RAHT transformation unit (7007), LOD generation unit (7008), inverse lifting unit (7009), and / or color inverse transformation unit (7010) can perform attribute decoding as described in FIG. 10. Attribute decoding according to the embodiments may include Region Adaptive Hierarchial Transform (RAHT) decoding, Interpolaration-based hierarchical nearest-neighbour prediction-Prediction Transform) decoding, and interpolation-based hierarchical nearest-neighbour prediction with an update / lifting step (Lifting Transform) decoding. The three decodings described above may be used optionally, or a combination of one or more decodings may be used. Furthermore, attribute decoding according to the embodiments is not limited to the examples described above.

[0153] The arismetic decoder (7005) according to the embodiments decodes the attribute bitstream into arismetic coding.

[0154] The inverse quantization unit (7006) according to the embodiments inverse quantizes information about the decoded attribute bitstream or the attribute obtained as a result of decoding and outputs the inverse quantized attributes (or attribute values). Inverse quantization may be optionally applied based on the attribute encoding of the point cloud encoder.

[0155] According to embodiments, the RAHT transformation unit (7007), LOD generation unit (7008), and / or inverse lifting unit (7009) may process the reconstructed geometry and inverse quantized attributes. As described above, the RAHT transformation unit (7007), LOD generation unit (7008), and / or inverse lifting unit (7009) may optionally perform a corresponding decoding operation according to the encoding of the point cloud encoder.

[0156] The color inverse conversion unit (7010) according to the embodiments performs inverse conversion coding to inversely convert the color value (or texture) included in the decoded attributes. The operation of the color inverse conversion unit (7010) may be selectively performed based on the operation of the color conversion unit (30006) of the point cloud encoder.

[0157] The elements of the point cloud decoder of FIG. 7 may be implemented in hardware, software, firmware, or a combination thereof, comprising one or more processors or integrated circuits configured to communicate with one or more memories included in the point cloud providing device, although not illustrated in the drawing. One or more processors may perform at least one of the operations and / or functions of the elements of the point cloud decoder of FIG. 7 described above. Additionally, one or more processors may operate or execute a set of software programs and / or instructions for performing the operations and / or functions of the elements of the point cloud decoder of FIG. 7.

[0158] FIG. 8 is an example of a transmission device according to embodiments.

[0159] The transmission device illustrated in FIG. 8 is an example of the transmission device (10000) of FIG. 1 (or the point cloud encoder of FIG. 3). The transmission device illustrated in FIG. 8 can perform at least one of the same or similar operations and methods as the operations and encoding methods of the point cloud encoder described in FIG. 1 to 6. A transmission device according to embodiments may include a data input unit (8000), a quantization processing unit (8001), a voxelization processing unit (8002), an octree occupancy code generation unit (8003), a surface model processing unit (8004), an intra / inter coding processing unit (8005), an arithmetic coder (8006), a metadata processing unit (8007), a color conversion processing unit (8008), an attribute conversion processing unit (or attribute conversion processing unit) (8009), a prediction / lifting / RAHT conversion processing unit (8010), an arithmetic coder (8011) and / or a transmission processing unit (8012).

[0160] The data input unit (8000) according to the embodiments receives or acquires point cloud data. The data input unit (8000) may perform an operation and / or acquisition method identical or similar to the operation and / or acquisition method of the point cloud video acquisition unit (10001) (or the acquisition process (20000) described in FIG. 2).

[0161] The data input unit (8000), quantization processing unit (8001), voxelization processing unit (8002), octree occupancy code generation unit (8003), surface model processing unit (8004), intra / inter coding processing unit (8005), and arithmetic coder (8006) perform geometry encoding. Since the geometry encoding according to the embodiments is identical or similar to the geometry encoding described in FIGS. 1 to 6, a detailed description is omitted.

[0162] The quantization processing unit (8001) according to the embodiments quantizes geometry (e.g., location values ​​of points, or position values). The operation and / or quantization of the quantization processing unit (8001) is the same or similar to the operation and / or quantization of the quantization unit (30001) described in FIG. 3. The specific description is the same as that described in FIG. 1 through 6.

[0163] The voxelization processing unit (8002) according to the embodiments voxelizes the position values ​​of the quantized points. The voxelization processing unit (80002) may perform the same or similar operation and / or process as the operation and / or voxelization process of the quantization unit (30001) described in FIG. 3. The specific description is the same as that described in FIG. 1 to 6.

[0164] The octree occupancy code generation unit (8003) according to the embodiments performs octree coding on the positions of voxelized points based on an octree structure. The octree occupancy code generation unit (8003) can generate an occupancy code. The octree occupancy code generation unit (8003) can perform operations and / or methods identical or similar to the operations and / or methods of the point cloud encoder (or octree analysis unit (30002)) described in FIGS. 3 and 4. The specific description is the same as that described in FIGS. 1 through 6.

[0165] The surface model processing unit (8004) according to the embodiments can perform trisup geometry encoding that reconstructs the positions of points within a specific region (or node) based on a voxel based on a surface model. The surface model processing unit (8004) can perform operations and / or methods identical or similar to the operations and / or methods of the point cloud encoder (e.g., surface approximation analysis unit (30003)) described in FIG. 3. The specific description is the same as that described in FIG. 1 through 6.

[0166] According to the embodiments, the intra / inter coding processing unit (8005) can intra / inter code point cloud data. The intra / inter coding processing unit (8005) can perform coding identical or similar to the intra / inter coding described in FIG. 7. The specific description is the same as that described in FIG. 7. According to the embodiments, the intra / inter coding processing unit (8005) may be included in an arismetic coder (8006).

[0167] An arismetic coder (8006) according to the embodiments entropy-encodes an octree and / or approximated octree of point cloud data. For example, the encoding method includes an arismetic encoding method. The arismetic coder (8006) performs the same or similar operation and / or method as the arismetic encoder (30004).

[0168] A metadata processing unit (8007) according to the embodiments processes metadata regarding point cloud data, such as setting values, and provides it to necessary processing processes such as geometry encoding and / or attribute encoding. Additionally, a metadata processing unit (8007) according to the embodiments may generate and / or process signaling information related to geometry encoding and / or attribute encoding. The signaling information according to the embodiments may be encoded separately from geometry encoding and / or attribute encoding. Additionally, the signaling information according to the embodiments may be interleaved.

[0169] The color conversion processing unit (8008), attribute conversion processing unit (8009), prediction / lifting / RAHT conversion processing unit (8010), and arithmetic coder (8011) perform attribute encoding. Since the attribute encoding according to the embodiments is identical or similar to the attribute encoding described in FIGS. 1 to 6, a detailed description is omitted.

[0170] The color conversion processing unit (8008) according to the embodiments performs color conversion coding that converts color values ​​included in attributes. The color conversion processing unit (8008) may perform color conversion coding based on reconstructed geometry. The description of the reconstructed geometry is the same as that described in FIGS. 1 through 6. In addition, it performs the same or similar operation and / or method as the operation and / or method of the color conversion unit (30006) described in FIG. 3. A detailed description is omitted.

[0171] The attribute transformation processing unit (8009) according to the embodiments performs attribute transformation that transforms attributes based on positions where geometry encoding has not been performed and / or reconstructed geometry. The attribute transformation processing unit (8009) performs operations and / or methods identical or similar to the operations and / or methods of the attribute transformation unit (30007) described in FIG. 3. A detailed description is omitted. The prediction / lifting / RAHT transformation processing unit (8010) according to the embodiments may code the transformed attributes by RAHT coding, prediction transformation coding, and lifting transformation coding, or a combination thereof. The prediction / lifting / RAHT transformation processing unit (8010) performs at least one of operations identical or similar to the operations of the RAHT transformation unit (30008), LOD generation unit (30009), and lifting transformation unit (30010) described in FIG. 3. In addition, the descriptions of predictive transformation coding, lifting transformation coding, and RAHT transformation coding are the same as those described in Figures 1 to 6, so a detailed description is omitted.

[0172] The arismetic coder (8011) according to the embodiments can encode coded attributes based on arismetic coding. The arismetic coder (8011) performs the same or similar operation and / or method as the operation and / or method of the arismetic encoder (300012).

[0173] A transmission processing unit (8012) according to embodiments may transmit each bitstream containing encoded geometry and / or encoded attributes and metadata information, or may transmit the encoded geometry and / or encoded attributes and metadata information by configuring them into a single bitstream. When the encoded geometry and / or encoded attributes and metadata information according to embodiments is configured into a single bitstream, the bitstream may include one or more sub-bitstreams. The bitstream according to embodiments may include signaling information and slice data, including SPS (Sequence Parameter Set) for sequence-level signaling, GPS (Geometry Parameter Set) for signaling of geometry information coding, APS (Attribute Parameter Set) for signaling of attribute information coding, and TPS (Tile Parameter Set) for tile-level signaling. The slice data may include information regarding one or more slices. One slice according to embodiments is one geometry bitstream (Geom0 0 ) and one or more attribute bitstreams (Attr0 0 , Attr1 0 It may include ).

[0174] A slice refers to a series of syntax elements representing all or part of a coded point cloud frame.

[0175] According to the embodiments, the TPS may include information regarding each tile (e.g., coordinate value information of a bounding box and height / size information, etc.) for one or more tiles. The geometry bitstream may include a header and a payload. The header of the geometry bitstream according to the embodiments may include identification information of a parameter set included in the GPS (geom_parameter_set_id), a tile identifier (geom_tile_id), a slice identifier (geom_slice_id), and information regarding data included in the payload, etc. As described above, the metadata processing unit (8007) according to the embodiments may generate and / or process signaling information and transmit it to the transmission processing unit (8012). According to the embodiments, the elements performing geometry encoding and the elements performing attribute encoding may share data / information with each other as indicated by the dotted lines. The transmission processing unit (8012) according to the embodiments may perform an operation and / or transmission method identical or similar to the operation and / or transmission method of the transmitter (10003). A detailed explanation is omitted as it is the same as that described in FIGS. 1 and 2.

[0176] FIG. 9 is an example of a receiving device according to embodiments.

[0177] The receiving device illustrated in FIG. 9 is an example of the receiving device (10004) of FIG. 1 (or the point cloud decoder of FIG. 10 and FIG. 11). The receiving device illustrated in FIG. 9 can perform at least one of the same or similar operations and methods as the operations and decoding methods of the point cloud decoder described in FIG. 1 to FIG. 11.

[0178] A receiving device according to the embodiments may include a receiving unit (9000), a receiving processing unit (9001), an arithmetic decoder (9002), an occupancy code-based octree reconstruction processing unit (9003), a surface model processing unit (triangle reconstruction, up-sampling, voxelization) (9004), an inverse quantization processing unit (9005), a metadata parser (9006), an arithmetic decoder (9007), an inverse quantization processing unit (9008), a prediction / lifting / RAHT inverse transformation processing unit (9009), a color inverse transformation processing unit (9010), and / or a renderer (9011). Each component of the decoding according to the embodiments may perform the inverse process of the components of the encoding according to the embodiments.

[0179] A receiver (9000) according to the embodiments receives point cloud data. The receiver (9000) may perform an operation and / or a receiving method identical or similar to the operation and / or receiving method of the receiver (10005) of FIG. 1. A detailed description is omitted.

[0180] A receiving processing unit (9001) according to the embodiments can obtain a geometry bitstream and / or an attribute bitstream from the received data. The receiving processing unit (9001) may be included in the receiving unit (9000).

[0181] The arismetic decoder (9002), the Occupancy code-based octree reconstruction processing unit (9003), the surface model processing unit (9004), and the inverse quantization processing unit (9005) can perform geometry decoding. Since the geometry decoding according to the embodiments is identical or similar to the geometry decoding described in FIGS. 1 to 10, a detailed description is omitted.

[0182] The arismetic decoder (9002) according to the embodiments can decode a geometry bitstream based on arismetic coding. The arismetic decoder (9002) performs the same or similar operation and / or coding as the operation and / or coding of the arismetic decoder (7000).

[0183] According to the embodiments, the Occupancy code-based octree reconstruction processing unit (9003) can reconstruct an octree by obtaining an Occupancy code from a decoded geometry bitstream (or information regarding geometry obtained as a result of decoding). The Occupancy code-based octree reconstruction processing unit (9003) performs the same or similar operations and / or methods as the octree synthesis unit (7001) and / or octree generation method. According to the embodiments, the surface model processing unit (9004) can perform trisup geometry decoding and related geometry reconstruction (e.g., triangle reconstruction, up-sampling, voxelization) based on the surface model method when trisup geometry encoding is applied. The surface model processing unit (9004) performs the same or similar operations as the surface offset synthesis unit (7002) and / or geometry reconstruction unit (7003).

[0184] The inverse quantization processing unit (9005) according to the embodiments can inverse quantize the decoded geometry.

[0185] A metadata parser (9006) according to the embodiments can parse metadata included in the received point cloud data, such as setting values, etc. The metadata parser (9006) can pass the metadata to geometry decoding and / or attribute decoding. A specific description of the metadata is omitted as it is the same as described in FIG. 8.

[0186] The arismetic decoder (9007), inverse quantization processing unit (9008), prediction / lifting / RAHT inverse transformation processing unit (9009), and color inverse transformation processing unit (9010) perform attribute decoding. Since attribute decoding is identical or similar to the attribute decoding described in FIGS. 1 to 10, a detailed description is omitted.

[0187] The arismetic decoder (9007) according to the embodiments can decode an attribute bitstream into arismetic coding. The arismetic decoder (9007) can perform decoding of the attribute bitstream based on reconstructed geometry. The arismetic decoder (9007) performs the same or similar operation and / or coding as the operation and / or coding of the arismetic decoder (7005).

[0188] The inverse quantization processing unit (9008) according to the embodiments can inverse quantize the decoded attribute bitstream. The inverse quantization processing unit (9008) performs the same or similar operation and / or method as the operation and / or inverse quantization method of the inverse quantization unit (7006).

[0189] According to the embodiments, the prediction / lifting / RAHT inverse transformation processing unit (9009) can process the reconstructed geometry and inverse quantized attributes. The prediction / lifting / RAHT inverse transformation processing unit (9009) performs at least one of the same or similar operations and / or decodings as the operations and / or decodings of the RAHT transformation unit (7007), LOD generation unit (7008), and / or inverse lifting unit (7009). According to the embodiments, the color inverse transformation processing unit (9010) performs inverse transformation coding to inversely transform the color values ​​(or textures) included in the decoded attributes. The color inverse transformation processing unit (9010) performs the same or similar operations and / or inverse transformation coding as the operations and / or inverse transformation coding of the color inverse transformation unit (7010). A renderer (9011) according to the embodiments can render point cloud data.

[0190] FIG. 10 shows an example of a structure that can be linked with a point cloud data transmission / reception method / device according to embodiments.

[0191] The structure of FIG. 10 represents a configuration in which at least one of a server (1060), a robot (1010), an autonomous vehicle (1020), an XR device (1030), a smartphone (1040), a home appliance (1050) and / or an HMD (1070) is connected to a cloud network (1010). The robot (1010), the autonomous vehicle (1020), the XR device (1030), the smartphone (1040), or the home appliance (1050) are referred to as devices. Additionally, the XR device (1030) may correspond to a point cloud data (PCC) device according to the embodiments or may be linked with a PCC device.

[0192] The cloud network (1000) may refer to a network that constitutes part of the cloud computing infrastructure or exists within the cloud computing infrastructure. Here, the cloud network (1000) may be configured using a 3G network, a 4G or LTE (Long Term Evolution) network, or a 5G network, etc.

[0193] The server (1060) is connected to at least one of a robot (1010), an autonomous vehicle (1020), an XR device (1030), a smartphone (1040), a home appliance (1050) and / or an HMD (1070) via a cloud network (1000) and can assist in at least some of the processing of the connected devices (1010 to 1070).

[0194] The HMD (Head-Mount Display) (1070) represents one of the types in which an XR device and / or PCC device according to the embodiments may be implemented. A device of the HMD type according to the embodiments includes a communication unit, a control unit, a memory unit, an I / O unit, a sensor unit, and a power supply unit, etc.

[0195] Hereinafter, various embodiments of the device (1010 to 1050) to which the above-described technology is applied are described. Here, the device (1010 to 1050) illustrated in FIG. 10 may be linked / coupled with a point cloud data transmission / reception device according to the above-described embodiments.

[0196] <PCC+XR>

[0197] The XR / PCC device (1030) may be implemented as a Head-Mount Display (HMD), a Head-Up Display (HUD) equipped in a vehicle, a television, a mobile phone, a smartphone, a computer, a wearable device, a home appliance, digital signage, a vehicle, a stationary robot, or a mobile robot by applying PCC and / or XR (AR+VR) technology.

[0198] The XR / PCC device (1030) can obtain information about surrounding space or real objects by analyzing 3D point cloud data or image data obtained through various sensors or from an external device to generate position data and attribute data for 3D points, and can render and output an XR object to be output. For example, the XR / PCC device (1030) can output an XR object containing additional information about a recognized object by associating it with the recognized object.

[0199] <PCC+XR+모바일폰>

[0200] The XR / PCC device (1030) can be implemented as a mobile phone (1040) or the like by applying PCC technology.

[0201] The mobile phone (1040) can decode and display point cloud content based on PCC technology.

[0202] <PCC+자율주행+XR>

[0203] The autonomous vehicle (1020) can be implemented as a mobile robot, vehicle, unmanned aerial vehicle, etc. by applying PCC technology and XR technology.

[0204] An autonomous vehicle (1020) equipped with XR / PCC technology may refer to an autonomous vehicle equipped with means for providing XR images, or an autonomous vehicle that is the subject of control / interaction within the XR images. In particular, the autonomous vehicle (1020) that is the subject of control / interaction within the XR images is distinguished from the XR device (1030) and can be interconnected with it.

[0205] An autonomous vehicle (1020) equipped with means for providing XR / PCC images can acquire sensor information from sensors including cameras and output XR / PCC images generated based on the acquired sensor information. For example, the autonomous vehicle (1020) can provide an XR / PCC object corresponding to a real object or an object in the screen to the occupant by providing an XR / PCC object by outputting an XR / PCC image with a HUD.

[0206] At this time, when the XR / PCC object is displayed on the HUD, at least a portion of the XR / PCC object may be displayed so as to overlap with the actual object to which the occupant's gaze is directed. On the other hand, when the XR / PCC object is displayed on a display provided inside the autonomous vehicle, at least a portion of the XR / PCC object may be displayed so as to overlap with an object on the screen. For example, the autonomous vehicle (1220) may display XR / PCC objects corresponding to objects such as lanes, other vehicles, traffic lights, traffic signs, motorcycles, pedestrians, buildings, etc.

[0207] VR (Virtual Reality) technology, AR (Augmented Reality) technology, MR (Mixed Reality) technology and / or PCC (Point Cloud Compression) technology according to the embodiments can be applied to various devices.

[0208] In other words, VR technology is a display technology that provides real-world objects or backgrounds solely as CG images. On the other hand, AR technology refers to a technology that displays virtual CG images alongside images of real objects. Furthermore, MR technology is similar to the aforementioned AR technology in that it mixes and combines virtual objects with the real world. However, it is distinguished from AR technology in that while AR technology maintains a clear distinction between real-world objects and virtual objects created from CG images, using virtual objects to complement real-world objects, MR technology regards virtual objects as having the same nature as real-world objects. To give a more specific example, the aforementioned MR technology is applied in hologram services.

[0209] However, recently, rather than clearly distinguishing between VR, AR, and MR technologies, they are also referred to as XR (extended Reality) technology. Therefore, embodiments of the present invention are applicable to all VR, AR, MR, and XR technologies. These technologies may utilize encoding / decoding based on PCC, V-PCC, and G-PCC technologies.

[0210] The PCC method / device according to the embodiments can be applied to a vehicle providing autonomous driving services.

[0211] Vehicles providing autonomous driving services are connected to PCC devices to enable wired / wireless communication.

[0212] When a point cloud data (PCC) transceiver according to the embodiments is connected to a vehicle for wired or wireless communication, it can receive and process content data related to AR / VR / PCC services that can be provided along with an autonomous driving service, and transmit it to the vehicle. Additionally, when the point cloud data transceiver is mounted on a vehicle, the point cloud transceiver can receive and process content data related to AR / VR / PCC services according to a user input signal received through a user interface device and provide it to the user. A vehicle or a user interface device according to the embodiments can receive a user input signal. The user input signal according to the embodiments may include a signal indicating an autonomous driving service.

[0213] The method / device according to the embodiments includes and can perform an effective LoD generation method for point cloud data compression.

[0214] The present embodiments describe a method for representing attributes based on an octree structure considered as the geometry structure of point cloud data. At this time, a method for matching point cloud data based on the location information of octree nodes (point paired octree: a method of matching attributes and positions to octree nodes) is discussed, and a method for efficiently coding attributes based on this is described. The present embodiments may be used when transmitting or outputting low-resolution or subsampled point cloud data using an octree node to which attributes are assigned.

[0215] The present embodiments deal with a technique for compressing data composed of point clouds. Specifically, they deal with a method for efficiently compressing attribute information based on a method of matching attribute information of a point cloud to an octree structure, and can be used to output data having variable resolution depending on the characteristics of a receiver or transmission path.

[0216] Point cloud data consists of the location of each data point (geometry: e.g., XYZ coordinates) and attributes (e.g., color, reflectance, intensity, grayscale, opacity, etc.). Point Cloud Compression (PCC) uses octree-based compression to efficiently compress distribution characteristics that are non-uniformly distributed in three-dimensional space, and compresses attribute information based on this.

[0217] The RAHT (region adaptive hierarchical transform) method is being considered as an attribute compression method for point clouds. The formula below is the basic formula for RAHT, representing the generation of x-1 level coefficients from the x-axis adjacency coefficients of level l using the transformation function T. Here, w_1 and w_2 constituting T represent the number of leaf nodes used to generate g_l, and the leaf node level represents the occupancy of each node.

[0218]

[0219] At this time, since the coefficients used in the operation depend on the octree structure, attribute encoding / decoding is possible only after the entire octree structure has been encoded / decoded, which can be a factor causing delay in systems requiring high-speed processing.

[0220] Another attribute coding method involves generating Levels of Detail (LoD) based on the distance of points and coding attributes sequentially. In this case, a process to find surrounding points must be performed beforehand. While the nearest neighbor search method, which is commonly used for this purpose, has the advantage of accurately finding surrounding points, it has the disadvantage of taking a long time to execute, which can cause delays in systems requiring high-speed processing.

[0221] The method / device according to the embodiments may include a block-based LoD generation method and a method for changing the sampling direction according to the LoD level for non-scalable cases as well.

[0222] The embodiments describe an improved coding method in relation to point cloud attribute coding. Modifications and combinations between the embodiments of the present invention are possible. The terms used in this document may be understood based on their intended meanings within the scope of their common usage in the field.

[0223] Encoder scheme: Point cloud data and octree matching, and attribute deduplication between nodes

[0224] When undergoing the octree colorization process, attribute or position information is matched to each octree node, allowing each level of the octree to function as low-resolution point cloud data. However, since attributes from parent nodes are selected from child nodes, information may be duplicated. In other words, due to recursive selection, there is a possibility that the same information will be duplicated even at higher levels; this can reduce coding efficiency as the total number of data points increases.

[0225] FIG. 11 illustrates a method for generating a colored octree with attribute duplication removed according to embodiments.

[0226] FIG. 11 is an example of a method for preventing the aforementioned problem. Referring to FIG. 11, octree colorization is performed to match the attributes and locations of point cloud data to octree nodes in the direction from the octree root node to the leaf node, and attribute duplication removal is performed to remove duplicate information between octree levels. Each step is applied on a neighbor basis within the octree level and can be performed recursively starting from the root level to the leaf level.

[0227] FIG. 11 is a flowchart illustrating a process of matching attributes to octree nodes while preventing duplication of point cloud data. An encoding method / device according to embodiments may include and perform the method of FIG. 11.

[0228] Referring to Fig. 11, the encoder receives point cloud data and an octree structure as input and performs leaf node-level colorization. Starting from the root node level, it performs neighbor detection within each level and undergoes a point data selection process suitable for that area. Based on the selected data, it then performs octree node colorization. Subsequently, it performs inter-node attribute duplication removal, which removes unnecessary information occurring between octree levels. Afterward, it checks whether processing for all nodes in the current level is complete (End of node in this level?). If it is not complete, it performs a neighbor search process; once the current level is complete, it steps down toward the octree leaf level and recursively repeats the process. And once all processing from the root level to the leaf level is completed, a colorized octree without duplication is generated.

[0229] (1) Leaf node level colorization

[0230] FIG. 12 illustrates a leaf node level colorization method according to embodiments.

[0231] FIG. 12 is a diagram for specifically explaining the leaf node level colorization method of FIG. 11.

[0232] For given point cloud data, point cloud attributes corresponding to each location are matched to leaf nodes in an octree structure. By utilizing the fact that the location information of occupied nodes at the octree leaf node level matches one-to-one with the location information of the actual point cloud data, attributes of the corresponding point cloud data (color, reflectance, etc.) can be matched to the octree leaf nodes that match each point cloud.

[0233] For example, when the point cloud data matching the occupied leaf nodes in the octree structure of FIG. 12 are denoted as d1 to d9, the position of each point can be defined as (0,2,0), (1,2,0), (0,3,0), (1,3,0), (3,2,2), (2,2,1), (3,2,3), (2,3,3), (3,3,3), and their attributes can be defined as c1 to c9. When the attribute value of the position (x,y,z) is defined as Attr(x, y, z), the attributes corresponding to the occupied leaf nodes (parts marked with 1) among the total 16 leaf nodes of the example can be represented as follows.

[0234] c1=Attr(0,2,0), c2=Attr(1,2,0), c3=Attr(0,3,0), c4=Attr(1,3,0),

[0235] c5=Attr(3,2,2), c6=Attr(2,2,1), c7=Attr(3,2,3), c8=Attr(2,3,3), c9=Attr(3,3,3)

[0236] Figure 12 illustrates the relationship between an octree structure and attribute matching at the leaf node level. Referring to Figure 12, the occupied nodes at the leaf node level, which is the lowest level of the octree structure, can correspond one-to-one with the location information of the actual point cloud data. In Figure 12, the number 1 indicates that the node is occupied, and 0 indicates an empty node. Attributes such as color or reflectance, which are unique properties of the point cloud data, can be assigned to each occupied leaf node. The symbols c1 through c9 at the bottom of Figure 12 represent the attribute values ​​matched to each node. When the 3D coordinates (x, y, z) of the point cloud data d1 through d9 are defined, the attribute value Attr(x, y, z) of the corresponding location corresponds to the attribute c_{n} value of each leaf node.

[0237] (2) Neighbor detection

[0238] FIG. 13 illustrates a neighbor search method according to embodiments.

[0239] Figure 13 is a diagram illustrating the neighbor search process of Figure 11.

[0240] Referring to Fig. 13, adjacent point clouds are found based on defined neighbors. For example, neighbors can be defined based on an octree structure, i.e., child nodes sharing a parent can be defined as neighbors; in this case, there is an advantage in that neighbor nodes can be found with a small amount of computation. If necessary, the range of parents determining neighbor nodes can be extended to multiple parent nodes, in which case information determining neighbors (the location of the parent nodes) can be adaptively transmitted (e.g., for each octree depth level).

[0241] Figure 13 explains the principle of defining neighbors and searching for adjacent nodes in an octree structure. Octree-based neighbors refer to child nodes that share the same parent node. In the tree graph on the left side of Figure 13, nodes at lower levels share a single parent node. The cube diagram on the right side of Figure 13 visualizes the process of the octree dividing a three-dimensional space into eight child nodes. Occupied nodes, marked with 1 in the tree structure, correspond to specific locations within this cube space where actual data exists. Additionally, the parent range determining peripheral nodes can be extended to multiple parent nodes as needed. This is useful when a wider range of adjacent data needs to be referenced. Furthermore, search efficiency can be optimized by adaptively transmitting the location of the parent node—which is information determining neighbors—at each octree depth level.

[0242] (3) Point data selection and octree node colorization

[0243] The point data selection and octree node colorization process of Fig. 11 is explained in more detail.

[0244] Point cloud data within the neighborhood can be selected as a method for assigning attributes to a parent node. In the present invention, point cloud data close to the position of the parent node can be selected as a selection criterion. Point cloud data at a position similar to the average of the positions of neighboring nodes can be selected as shown below, and selection can be made in different ways depending on the application. The selected point - Attr(x_n, y_n, z_n) can be matched to an octree node by matching it to a target node position (x, y, z).

[0245]

[0246] In this case, the definition of the neighbor selected as information matching the parent node can be additionally defined as an 'unselected child node (or descendant node)', and one of the matching point cloud data can be selected according to the above formula. At this time, the selected point cloud data can be set to s() = 0 to prevent duplicate selection when matching other nodes. Alternatively, a matching node can be selected through other methods; for example, the median can be used as an approximation of the node's location. Or, a location corresponding to a specified order within the neighborhood can be used. If different types of selection methods can be used, each selection method can be signaled at the sequence level, octree depth level, or neighbor level.

[0247]

[0248] To reiterate the point data selection and octree node colorization process, when assigning attributes to a parent node, the point cloud data closest to that parent node's position can be selected. Specifically, data located at a position similar to the average value of the neighboring nodes' positions is selected. The attributes of the octree node can be determined by matching this selected point Attr(x_n, y_n, z_n) to the target node position (x, y, z). Furthermore, when determining the data to be matched to the parent node, the range of neighbors can be defined as 'child nodes (or descendant nodes) that have not yet been selected.' When specific point cloud data is selected, s() is set to 0 to prevent duplicate selection when matching with other nodes. Additionally, depending on the characteristics of the application, various approximation methods can be used in addition to the average value. For example, the median can be used as an approximation for the node position. Alternatively, data at a position corresponding to a predefined order within the neighbors can be used. When different selection methods are used depending on the situation, control information can be transmitted to efficiently communicate this. This signaling can be performed at the sequence level, octree depth level, or neighbor level.

[0249] (4) Duplicate attribute removal

[0250] FIG. 14 illustrates a method for removing duplicate attributes according to embodiments.

[0251] FIG. 14 is a diagram for specifically explaining the method of removing duplicate inter-node attributes of FIG. 11.

[0252] When matching attribute / position information to each octree node through the octree colorization process, information may be duplicated overall because attributes of a parent node are selected from its child nodes. Furthermore, since attribute selection is performed recursively according to the octree level, there is a possibility that the same information will be duplicated at higher levels; this can be a factor that reduces coding efficiency as the total amount of data increases.

[0253] To prevent this, the point cloud data selected during the attribute selection process can be removed as shown in FIG. 14, thereby preventing duplicate information from being transmitted during attribute coding. In this case, if point cloud data that was not previously selected is selected and matched in the process of (3), duplicate information in the leaf nodes (e.g., leaf nodes that match point cloud data set to s() = 0) can be removed after octree colorization. The duplicated attribute removal process can be performed after the octree colorization process is completed, or the removal of selected point cloud data can be performed immediately after the selection process or at each octree level as shown in FIG. 14.

[0254] Figure 14 shows a step-by-step attribute removal process to prevent information duplication and increase coding efficiency during the octree colorization process.

[0255] Step 1 represents the process of matching point cloud data to leaf nodes. Input data c_1 through c_9 are matched one-to-one with the leaf nodes of the octree. At this stage, the attribute information of the corresponding point cloud is assigned to every leaf node location.

[0256] Step 2 represents the selection and matching process at the root and leaf-1 level. Representative attributes of the parent level nodes are selected from the child nodes through a recursive process. At the root level, c_2 among the child nodes is selected and matched as the representative attribute of the root node. Then, at the next level (leaf-1) nodes, c_1 and c_5 are selected as their respective parent node attributes. At this point, the selected data (c_1, c_2, c_5) exists redundantly in the parent nodes, which can increase the total amount of data and reduce coding efficiency.

[0257] Step 3 represents the duplicated attribute removal process. Point cloud data already selected at the upper level is removed at the leaf node level. Looking at the final step (Points to be encoded) in Fig. 14, it can be seen that c_1, c_2, and c_5, which were matched at the root and (leaf-1) levels, have been deleted (indicated by '-') at the leaf node location. This allows control to prevent the transmission of duplicate information during attribute coding. This removal process can be performed after the entire octree colorization process is completed, or it can be performed immediately after selection at each level, as shown in Fig. 14. When selecting and matching previously unselected data, duplicate information at the leaf node can be efficiently removed by utilizing a flag indicating selection status (e.g., s() = 0).

[0258] Decoder Operations: Attribute Reconstruction and Octree-to-Point Cloud Data Matching

[0259] FIG. 15 illustrates the decoder operation according to the embodiments.

[0260] When point cloud data is compressed using the proposed method, the receiver requires a process to match the attribute data transmitted after decoding with the point cloud data.

[0261] Referring to FIG. 15, FIG. 15 is a flowchart illustrating the entire process of reconstructing and matching point cloud data and its attributes at the receiving end. First, the decoder can perform the attribute reconstruction and node matching process. The decoder can restore attribute information by receiving decoded attributes and an octree structure as input. Then, attributes can be reconstructed by descending from the root to the leaf level and using the attributes of the parent nodes as the predicted values ​​of the child nodes. At this time, a method of adding residual errors, similar to that of the transmitting end, can be used. The restored attribute information can then be matched with octree nodes (Attribute to node matching), and depending on the application, the result of this stage can be utilized as a low-resolution point cloud. Finally, the decoder can perform the position estimation and point matching process. The decoder can enter the stage of estimating actual point locations, rather than simple node locations, if full-level information is available by checking the flag (octree_full_level_present_flag == 1). Based on the octree structure and neighbor definitions, the decoder can estimate the location information of the actual point cloud. To prevent duplicate selections, selected data can be managed by setting s() = 0. Subsequently, based on the estimated location information, the decoder can precisely match the attributes that were matched to the nodes to the actual point locations. The decoder then repeats the process until all nodes within each level are processed, and performs the process by stepping down until it reaches the leaf level.

[0262] 1) Attribute reconstruction and attribute-to-node matching

[0263] FIG. 16 illustrates the process of matching attributes to octree nodes according to the embodiments.

[0264] FIG. 16 illustrates the process of reconstructing attribute information and matching it to an octree structure according to the present invention (attribute-to-node matching: the process of matching to an octree node at each level).

[0265] The receiver can detect peripheral nodes based on an octree structure for reconstructing location information in the same way as the transmitter, and can define peripheral nodes as sibling nodes having the same parent. If a different definition (e.g., a different size) is used for peripheral nodes, the receiver can transmit information for detecting peripheral nodes.

[0266] In the receiver, attributes can be predicted as the level descends (in the direction from the root to the leaf, as shown in the example below), similar to attribute prediction in the transmitter. The prediction technique can use the same method as used in the transmitter, and as in the embodiment of the present invention, the reconstructed attribute of the parent node can be used as the prediction value for the child. In this case, the range to which the prediction value is applied can be applied differently depending on the definition of a neighbor.

[0267]

[0268] To reconstruct the attributes of each child node based on predicted attribute information, the receiver can perform the inverse of the prediction error generation method used by the transmitter. For example, if a prediction error is generated based on the difference between the original attribute and the predicted attribute, the attribute information can be reconstructed by adding the estimated predicted attribute and the decoded prediction error as shown below. If there are various prediction error generation methods, information regarding the method to be used by the receiver can be transmitted.

[0269]

[0270] The attribute information restored through the above process can be considered as information where attribute data and octree nodes are matched. In this case, depending on the application, the attribute data matched with octree node positions can be used as low-resolution point cloud data.

[0271] Figure 16 illustrates the process of restoring attributes step-by-step and matching them to nodes using data decoded at the receiving end and an octree structure. First, the octree attribute reconstruction and node matching process is described. The decoder can receive the decoded representative attribute (c_2) and the remaining prediction errors (c'_1, c'_3, …, c'_9) as input. It can start with the occupied node locations identified in the initial state of the octree structure. Then, the restored attribute c_2 can be matched to the root node, which is the highest level. At this stage, the remaining data remains in the form of prediction errors (residuals). Subsequently, at the upper level (leaf - 1 node level), attribute prediction and reconstruction can be performed by descending levels from the root toward the leaf. The restored attribute (c_2) of the parent node can be used as the prediction value for the child node. The attributes (c_1, c_5) at the corresponding level can be restored by adding the decoded prediction error (c'_1, c'_5) to this predicted value. The result of this step serves as a partial representation and can be utilized as low-resolution point cloud data. The decoder then reaches the lowest level, the leaf node, to restore all attributes (c_3, c_4, c_6, c_7, c_8, c_9). Information can be obtained where the location of each node is completely matched with the restored attribute data. The decoder determines the prediction range using the same neighbor definition as the transmitter and, if necessary, can receive and use additional information for detecting surrounding nodes. The prediction error generated by the difference between the original attributes and the predicted attributes can be processed in reverse. That is, the final attribute information can be reconstructed by adding the decoded prediction error to the estimated predicted attributes.

[0272] 2) Position estimation and octree-to-point cloud data matching

[0273] If actual position information can be estimated based on the attributes of the reconstructed point cloud data, it is possible to output point cloud data at actual locations rather than arbitrary locations (for example, node positions are location information different from the actual location in the point cloud data), even when outputting results at intermediate octree depth levels through scalable decoding / representation. In this case, when performing scalable representation using attributes matched to arbitrary octree depth levels, it is possible to represent the entire point cloud data as a subset.

[0274] To estimate the location information of the actual matching point cloud data for a point paired node, the colorization process performed at the transmitting end can be carried out. For example, when neighbors are defined based on an octree structure and there is location information of point cloud data (or occupancy information of the leaf node) corresponding to an octree leaf node, the location information of the point cloud data corresponding to the octree node can be obtained as follows, as in the example. If selection is made at the transmitting end based on a different criterion / method, the receiving end can also perform selection based on the same criterion / method based on the signaling regarding this.

[0275]

[0276] At this time, in addition to the definition of neighbors selected as information matching the parent node, a child node (or descendant node) that has not yet been selected can be defined, and one of the matching point cloud data can be selected according to the above formula. At this time, the selected point cloud data can be set to s() = 0 to prevent duplicate selection when matching other nodes.

[0277] At this point, by storing the location matching information of the selected point cloud data for each node position in a separate buffer, the restored attributes for each node position can be matched to the actual point cloud data positions. (In the figure, this is illustrated as the process of matching to leaf node positions.)

[0278]

[0279] Here It maps node locations to point cloud data locations, and represents point cloud data and the location of the node, respectively.

[0280] FIG. 17 illustrates the attribute-node matching and location estimation process according to the embodiments.

[0281] FIG. 17 illustrates, in addition to the process of reconstructing attribute information and matching it to an octree structure (attribute-to-node matching: a process of matching to an octree node at each level), the process of estimating position information (position estimation: a process of estimating the position and matching it to a leaf node) according to the embodiments.

[0282] Attribute-to-node matching is the process of matching restored attributes with octree nodes; by transmitting the attribute bitstreams in octree breadth-first and Morton code order, the reconstruction results can be directly matched with octree nodes. Alternatively, the location of the node to which each attribute matches can be transmitted via a separate signal.

[0283] Point cloud data corresponding to attributes matched to octree nodes can be restored through position estimation for attributes matched to nodes. In other words, attribute-position pairing can be performed by finding the location (leaf node location in the figure above) that matches each octree node. By recursively performing this process for each octree depth level, point cloud data matched to each node can be found, and the entire point cloud data can be restored by matching the remaining information to unmatched leaf nodes in order.

[0284] The recovered data may be output in different forms depending on the receiver's performance or the type of information transmitted.

[0285] (case 1): If all octree full-level and attribute-related information is transmitted, the attribute-to-point matching described above is possible.

[0286] (case 2): If a partial octree level is transmitted, or if only a portion of the octree is restored due to receiver performance, attribute-point matching is difficult, so attributes matched to the octree nodes can be output. In this case, it can be seen that point cloud data for approximate locations is output.

[0287] (case 3): If the entire octree level is passed but only a part of the attribute is passed or only a part is output, the attribute-point matching described above is performed only up to some octree levels. By outputting all points matched so far, not only can a larger number of points be output compared to case 2, but a subset of point cloud data reflecting actual locations can also be output.

[0288] Figure 17 illustrates various cases in which attributes restored from an octree structure are matched with actual location information in stages, and data is output according to the receiving environment. The attribute-node matching process is a process of directly connecting the restored attribute information to an octree node. By transmitting the attribute bitstream in breadth-first and Morton code order, immediate matching with a node can be achieved without separate computation. In Figure 17, this can be verified through the process in which c_2 is matched to the root node. The position estimation and pairing process is a process of finding the actual point cloud location based on the attributes matched to the node. Through this, attribute-position pairing is performed, and the entire data, including unmatched leaf nodes, can be restored by recursively executing it according to the octree depth level. There may be various output cases depending on receiver performance and data availability.

[0289] (Case 1): Full representation—This is the result when all octree levels and attribute information are available. As shown at the bottom of the figure, the entire point cloud data can be restored and output with precise matching from c_1 to c_9 to all leaf node locations.

[0290] (Case 2): Partial representation for level n-1 - This is a case where only some levels of the octree are transmitted or only the upper levels are restored due to receiver performance limitations. Since attribute-point matching is not perfect, the attributes matched to the octree node locations are output as is, which can be viewed as approximate data for the actual locations.

[0291] (Case 3): Partial representation with real position - This is the case where the entire octree level is known but only partial attribute information is available. It outputs only the points that have been matched so far, and can show more points than Case 2, as well as output a subset of data that reflects actual position information.

[0292] 2) Syntax and Semantics

[0293] FIG. 18 shows attribute parameter set syntax according to embodiments.

[0294] FIG. 19 shows the general attribute bitstream and attribute slice header syntax according to the embodiments.

[0295] FIG. 20 shows attribute slice data syntax according to embodiments.

[0296] Signaling information according to the embodiments can be used at a transmitting end or a receiving end, etc. As an embodiment of the present invention, information regarding a colorized octree can be defined in an attribute parameter set and an attribute slice as follows, and depending on the application or system, it can be defined in a corresponding location or a separate location to use different application scopes and application methods. In this embodiment, the information is described as being defined independently of the attribute coding technique, but it can be defined in conjunction with the attribute coding method and can be defined in a geometry parameter set for geometry scalability. In addition, if the syntax element defined below can be applied to multiple point cloud data streams as well as the current point cloud data stream, it can be transmitted through a higher-level parameter set (e.g., sequence parameter set, tile parameter set, etc.).

[0297] Octree full level present flag (octree_full_level_present_flag): If 1, it indicates that the octree full level is transmitted, and if 0, it indicates that a partial octree is transmitted. When octree_full_level_present_flag is 1, partial / full point cloud data can be output by performing attribute-to-node matching and attribute-to-point matching processes through the method proposed in the present invention. When octree_full_level_present_flag is 0, a partial octree containing attributes can be output through the attribute-to-node matching process.

[0298] scalable_representation_available_flag: If 1, it indicates that the decoded PCC data is a structure that considers a scalable representation. If 0, it indicates that the decoded PCC data is not a structure that considers a scalable representation. In this case, the receiver can perform octree colorization for a scalable representation.

[0299] octree_colorization_type: Can indicate the method used for octree colorization. If 0, it indicates that an attribute paired octree was used, and if 1, a point paired octree was used.

[0300] matched_attribute_type: Indicates the characteristics of the attribute matched with the octree node. If 0, it may represent an estimated attribute value (e.g., estimated based on attributes of neighboring nodes or child nodes), and if 1, it may represent an actual attribute value (e.g., estimated representative of attributes matched to child nodes).

[0301] attribute_selection_type: Indicates how attributes are matched to octree nodes. If 0, it indicates the average value; if 1, the median value; and if 2, a selection method from fixed positions (e.g., first, second, etc.).

[0302] point_data_selection_type: This indicates the method of selecting point cloud data when matching actual point cloud data (attributes and geometry) among the methods of matching point data to an octree node. If 0, it indicates a fixed location; if 1, it indicates actual point cloud data close to the location average; if 2 (corrected typo for 3 in original text), it indicates a middle location (e.g., the median value in Morton code); and if 3, it indicates a method of indicating the actual location.

[0303] point_cloud_geometry_info_present_flag: If 1, it indicates whether geometry information for point cloud data matched with an octree node is provided directly.

[0304] colorization_start_depth_level: Indicates the starting octree level with octree colorization applied.

[0305] colorization_end_depth_level: Indicates the end octree level to which octree colorization is applied.

[0306] num_colorized_nodes[i]: Represents the number of nodes to which colorization has been applied for the (i + colorization_start_depth_level)th octree depth level.

[0307] position_index[i][j]: Can represent the position information of the j-th node for the (i + colorization_start_depth_level)-th octree depth level. You can directly pass XYZ position information, pass information that has been transformed into XYZ positions such as Morton code, or pass an index that matches the position (e.g., the order when child nodes sharing a parent node are sorted in Morton code order (1-8)).

[0308] The abh_attr_parameter_set_id within attribute_slice_header() can be matched with the aps_attr_parameter_set_id in attribute_parameter_set(). This allows determining which colorization method or deduplication rule a specific data slice follows.

[0309] octree_full_level_present_flag is a flag that determines whether the decoder operates in 'precise restoration' mode or 'approximate restoration' mode. When this flag is 1, attribute-point matching performed implies restoration to the actual coordinate system, so although the amount of computation increases, high-quality output such as Case 1 is possible.

[0310] 3) Flowchart for Implementation of Example (End-to-End System Perspective)

[0311] 1) Encoder, Transmitter

[0312] FIG. 21 shows the encoder configuration according to the embodiments.

[0313] FIG. 21 is an example of a detailed functional configuration for encoding / transmitting PCC data using the embodiments thereof. When point cloud data is input, the encoder can process geometry data (e.g., XYZ coordinates, phi-theta coordinates, etc.) and attribute data (e.g., color, reflectance, intensity, grayscale, opacity, medium, material, glossiness, etc.).

[0314] At this time, the present invention describes a technology for processing attribute information, specifically comprising (1) a part that matches attributes and positions of point cloud data to an octree node (point cloud data to octree matching), (2) a part that removes point cloud data that is duplicated between octree levels (inter-level attribute duplication removal), (3) a part that estimates and removes similarity between attribute information, (4) a part that converts the estimated information into a format suitable for transmission or a domain with high compression efficiency and quantizes it, and (5) entropy coding that converts it into bit-unit data for transmission.

[0315] At this time, (1) the attribute estimation block can be performed based on the spatial distribution characteristics of adjacent data. Also, (2) the transformation block can use various transformation techniques (e.g., DCT series transformation, lifting transform, RAHT, wavelet transform series, etc.) depending on the type of data, and the data can be transmitted after quantization without transformation.

[0316] (1) Detection and prediction of peripheral nodes: As a method for detecting peripheral nodes, prediction can be performed based on the positional proximity between child nodes within a certain range. For example, in an octree structure, it can be assumed that child nodes originating from the same parent node are adjacent to each other, and that the predicted values ​​between adjacent child nodes are similar. Accordingly, in the present invention, instead of calculating a predicted value for each child node individually, sibling nodes having the same parent node are defined to have the same predicted value. This can help increase coding efficiency by reducing the number of coefficients required when encoding each child node. Additionally, by acting as a representative value for each parent node, attributes matching the octree structure can be predicted.

[0317]

[0318] (2) Generation of prediction error information: Based on the prediction information, the attribute prediction error of each child node can be calculated as follows. In the present invention, the difference between the original attribute value and the predicted attribute value is used, but other methods (e.g., weighted difference, weighted averaged difference, etc.) may be used depending on the method of implementation or purpose.

[0319]

[0320] (3) Information Selection: According to the present invention, nodes at lower levels (in the direction of leaf nodes) can be defined to have a large level value ($n=N$), while nodes at higher levels (root nodes) can have a small level value ($n=0$). (In some cases, the reverse order is also possible.) At this time, the type of information transmitted differs depending on each level; while the highest level transmits prediction attribute values, the lower levels transmit prediction errors.

[0321] (4) Bit delivery: The delivery order of data can be determined by level, taking into account the decoder processing process. It can be delivered in increasing order of levels, starting with the predicted attribute value of the upper level, and within each level, in increasing order along the XYZ axes (e.g., Morton code order). If necessary, a reordering process can be added.

[0322] (5) Quantization: The following quantization process can be applied to the predicted value and the prediction error. The degree of quantization is determined by the quantization coefficient (q), and different quantization coefficients can be used depending on the predicted value and the prediction error, or depending on the level.

[0323]

[0324] FIG. 21 shows a detailed functional block diagram of an encoder for encoding and transmitting PCC data. Regarding the geometry data processing (Geometry Path), the location information of the input point cloud is converted into a hierarchical structure through an octree generation step. Subsequently, it can be generated into a final geometry bitstream through geometry prediction and entropy coding. The octree structure information generated at this time can be passed to the attribute processing path for reference. Regarding the attribute data processing (Attribute Path), during the point cloud data-octree matching process, colorization can be performed to match the attributes and locations of the point cloud to the octree nodes. During the inter-node duplication removal process, information existing redundantly between octree levels can be removed to improve coding efficiency. During the attribute prediction process, a predicted value can be generated based on the similarity between sibling nodes that share the same parent. The number of coefficients to be transmitted can be reduced by defining a common prediction value instead of calculating individual prediction values ​​for all child nodes. Furthermore, during the Transform & Quantization process, prediction error (residual) information can be transformed into a domain with high compression efficiency and quantized using DCT, RAHT, lifting transforms, etc. Quantization coefficients can be applied differently depending on the level or data characteristics. Finally, during the Entropy Coding process, the compressed data can be converted into a bit-unit attribute bitstream.

[0325] 2) Decoder / Receiver

[0326] FIG. 22 shows a decoder configuration according to embodiments.

[0327] The decoding method and apparatus according to the embodiments (receiving device (10004) of FIG. 1, receiving (20002) to rendering (20004) of FIG. 2, decoder of FIG. 7 and FIG. 9, each device of FIG. 10, decoder of FIG. 22, bitstream and parameter acquisition of FIG. 18 to FIG. 20, bitstream and parameter acquisition of FIG. 24 to FIG. 25, decoding of FIG. 27, method of FIG. 29) may be configured as shown in FIG. 22.

[0328] FIG. 22 is an example of a detailed functional configuration for receiving / decoding PCC data using the present invention. When a bitstream is input, the receiver can process the bitstream for position information and the bitstream for attribute information separately. The bitstream for geometry information may undergo a process of regenerating an octree after entropy decoding. The attribute bitstream can undergo (1) entropy decoding, (2) dequantization, and (3) inverse transform processes, and then (4) attribute information prediction and attribute regeneration (attribute regeneration) of attributes corresponding to octree nodes, and (5) a process of restoring leaf node level information based on the restored attribute information, and (6) a process of outputting point cloud data at various resolutions depending on receiver performance or application.

[0329] In this case, if the colorized octree itself possesses both attribute and position information, the output of octree leaf level colorization can be used as the final output without a separate output process for geometry. In this case, various techniques may be used for the inverse quantization and inverse transform processes depending on the quantization and transform processes used by the transmitter, and if the data is encoded without a transform or quantization process, the inverse transform or inverse quantization process may not be used.

[0330] (1) Peripheral node detection

[0331] The receiver can detect peripheral nodes based on an octree structure for reconstructing location information in the same way as the transmitter, and can define peripheral nodes as sibling nodes having the same parent. If a different definition (e.g., a different size) is used for peripheral nodes, the receiver using it can transmit information for detecting peripheral nodes.

[0332] (2) Attribute prediction based on peripheral node information

[0333] The receiver predicts attributes in the reverse order of the sender's attribute prediction, descending the levels (from the root to the leaf). The prediction technique uses the same method as that used by the sender, and the reconstructed attribute of the parent node can be used as the prediction value for the child. If multiple methods can be used, the method used during encoding and additional information can be transmitted.

[0334]

[0335] (3) Reconfiguration of attributes

[0336] To reconstruct the attributes of each child node based on the predicted attribute information at the receiver, the inverse of the prediction error generation method used by the transmitter is performed. For example, if a prediction error is generated based on the difference between the original attribute and the predicted attribute, the attribute information can be reconstructed by adding the estimated predicted attribute and the decoded prediction error as shown below. Information regarding the method to be used by the receiver, given that there are various prediction error generation methods available, can be conveyed.

[0337]

[0338] (4) Information output

[0339] Generally, the goal is for the attributes of a leaf node to output the final result, but depending on receiver performance, application field, availability of information, etc., the result may be output at a certain level. In this case, the level to be output and related information can be transmitted.

[0340] FIG. 22 shows a detailed functional block diagram of a decoder that receives PCC data and restores geometry and attributes. The flow of reconstructing a point cloud through the reverse process of the transmitting end (encoder) can be described as follows.

[0341] Regarding the geometry path, the geometry portion is separated from the input bitstream and entropy decoding is performed. Subsequently, the octree structure, which is the positional system of the points, can be restored through the octree reconstruction process. This structural information is used as key reference data in the attribute restoration process.

[0342] Regarding the attribute path, during the initial decoding process, data in the transformation domain can be converted into a reconstructed res. / pred. attribute form through entropy decoding, inverse quantization, and inverse transform. Then, during the attribute reconstruction process, peripheral nodes are detected by referencing the reconstructed octree structure. The attributes of each node can be reconstructed by using the reconstructed attribute value of the parent node as the prediction value of the child node and adding the decoded prediction error to it. Subsequently, during the octree leaf-level colorization process, the final leaf node-level color and attribute information can be restored based on the reconstructed attribute information. Finally, during the scalable representation process, point cloud data of various resolutions can be output by selecting a target resolution (representation depth level) depending on the receiver's performance or application field.

[0343] Modifications and combinations between the embodiments are possible. Terms used in the embodiments may be understood based on their intended meanings within the scope of their widespread use in the field. Examples described in the embodiments may be considered together with the embodiments described above.

[0344] When considering partial decoding, such as with Fine Granularity Scalability (FGS), the geometry tree structure and the level of detail (LoD) structure need to be aligned. This is because, if the depth of the occupancy tree matches the level of detail (LoD), partial decoding can be performed by skipping the corresponding level of detail when skipping one depth of the occupancy tree. However, to achieve this, the occupancy tree structure must be considered when generating the level of detail, and in FGS, a level of detail corresponding to each depth of the occupancy tree can be generated through block-based subsampling.

[0345] It is specified that in FGS attribute decoding, a block-based subsampling method is used for detail level generation.

[0346] Creation of a single detail level

[0347] When fgs_layer_group_enabled is equal to 1, block-based subsampling (10.6.5.8) is performed.

[0348] However, block-based subsampling (10.6.5.8) does not describe the detailed conditions for FGS attribute decoding where the detail level structure must be matched with the occupancy tree.

[0349] Levels of detail

[0350] When fgs_layer_group_enabled is 1, the general detail level generation process is specified by E.6.3.3.1.

[0351] fgs_layer_group_enabled being equal to 1 specifies that the slice consists of multiple fine granularity slices (FGS) of partial slice geometry or partial slice attributes. fgs_layer_group_enabled being equal to 0 specifies that the slice does not consist of FGS. When fgs_layer_group_enabled is equal to 1, the partially decoded occupancy tree can be reconstructed as specified in Annex E. When fgs_layer_group_enabled does not exist, fgs_layer_group_enabled is inferred to be 0.

[0352] The requirement for bitstream conformance is that fgs_layer_group_enabled must be 0 if any of the following conditions apply.

[0353] geom_tree_type is 1 or,

[0354] If occtree_coded_axis_list_present is 1, or

[0355] If geom_scaling_enabled is 1 and geom_qp_mul_log2 is not 3,

[0356] geom_angular_enabled is 1 or,

[0357] When inter_prediction_enabled is 1.

[0358] General generation process

[0359] When fgs_layer_group_enabled is 1, the minimum level of detail is specified in the variable LodMinLevel. The maximum level of detail is set to overlap with the minimum level of detail of the parent FGS.

[0360] The finest level of detail is identified by the detail level index LodMinLevel. LodMinLevel is set to the log2 quantized node size of the leaf nodes of the FGS geometry.

[0361] The detail level is iteratively subsampled (E.6.3.3.4) starting from the finest detail level until only one point remains or LodMaxLevel subsampled detail levels are generated. The variable Lvl identifies the detail level to be subsampled. When layer_group_id is 0, LodMaxLevel is set to the log2 quantized node size of the root nodes of the FGS geometry. Otherwise, LodMaxLevel is set to the log2 quantized node size of the root nodes of the FGS geometry plus 1.

[0362] LodMinLevel = occtreeMaxDepthMinus1 - (startDepth-1)

[0363] LodMaxLevel = LodMinLevel+num_layers_minus1[layer_group_id]+1

[0364] if(layer_group_id>0)

[0365] LodMaxLevel++

[0366] Lvl = LodMinLevel

[0367] for (; Lvl < LodMaxLevel; Lvl++) {

[0368] if (LodPtCnt[Lvl] == 1)

[0369] break

[0370] … / * LodPtIdx[Lvl] subsampling * /

[0371] }

[0372] LodCnt = Lvl + 1

[0373] The coarsest detail level is identified by the detail level index LodCnt - 1. All points of the coarsest detail level are assigned to the coarsest level refinement list (10.6.5.3).

[0374] The finest detail level

[0375] The AttrPos point indices of the finest detail level have an initial one-to-one correspondence with the canonical decoding order of the FGS geometry.

[0376] subgroupPointCnt = SubgroupNodeCnt[layerGroupIdx][subgroupIdx] + SubgroupDirectNodePointCnt[layerGroupIdx][subgroupIdx]

[0377] for (ptIdx = 0; ptIdx < subgroupPointCnt; ptIdx++)

[0378] LodPtIdx[LodMinLevel][ptIdx] = ptIdx

[0379] LodPtCnt[LodMinLevel] = subgroupPointCnt

[0380] Point indices at the finest level of detail are sorted by group in ascending order of their respective Morton-coded attribute coordinates. The variable maxPtsPerSort identifies the maximum group size when sorting by group.

[0381] maxPtsPerSort = !attr_canonical_order_enabled && !max_points_per_sort_log2_plus1 ? LodPtCnt[LodMinLevel] : 1 << (max_points_per_sort_log2_plus1 - 1)

[0382] The sorted order must be the same for all attribute decodings within the FGS that have the same attribute coordinate array (AttrPos).

[0383] Performing a stable sort on each attribute or reusing the reordered points satisfies the requirement for the same order.

[0384] An exemplary sorting process is as follows.

[0385] for (benIdx = 0; benIdx < LodPtCnt[0]; benIdx += maxPtsPerSort) {

[0386] endIdx = Min(benIdx + maxPtsPerSort, LodPtCnt[LodMinLevel])

[0387] for (i = benIdx; i < endIdx; i++)

[0388] for (j = i + 1; j < endIdx; j++) {

[0389] iPtIdx = LodPtIdx[LodMinLevel][i]

[0390] jPtIdx = LodPtIdx[LodMinLevel][j]

[0391] iMorton = Morton(AttrPos[iPtIdx][0], AttrPos[iPtIdx][1], AttrPos[iPtIdx][2])

[0392] jMorton = Morton(AttrPos[jPtIdx][0], AttrPos[jPtIdx][1], AttrPos[jPtIdx][2])

[0393] if (iMorton > jMorton)

[0394] Swap(LodPtIdx[LodMinLevel][i], LodPtIdx[LodMinLevel][j])

[0395] }

[0396] }

[0397] The finest detail level of the parent FGS

[0398] The expression RefAttrPos [ptIdx][k] specifies the coordinates of each point for attribute coding in the parent FGS. The RefAttrPos point indices at the finest level of detail have an initial one-to-one correspondence with the standard decoding order of the parent FGS geometry. The variable parentPointCnt is the size of the number of points in the parent FGS.

[0399] for (ptIdx = 0; ptIdx < parentPointCnt; ptIdx++)

[0400] LodPtIdx[ptIdx] = ptIdx

[0401] parentLodPtCnt = parentPointCnt

[0402] Point indices of the finest detail levels are sorted by group in ascending order of their respective Morton-coded attribute coordinates. The sorted order must be the same for all attribute decodings within a single FGS that have the same attribute coordinate array (RefAttrPos).

[0403] Performing stable sorting for each attribute or reusing reordered points satisfies the requirement for the same order.

[0404] An exemplary sorting process is as follows.

[0405] for (benIdx = 0; benIdx < parentPointCnt; benIdx += maxPtsPerSort) {

[0406] endIdx = Min(benIdx + maxPtsPerSort, parentPointCnt)

[0407] for (i = benIdx; i < endIdx; i++)

[0408] for (j = i + 1; j < endIdx; j++) {

[0409] iPtIdx = ParentLodPtIdx[i]

[0410] jPtIdx = ParentLodPtIdx[j]

[0411] iMorton = Morton(RefAttrPos[iPtIdx][0], RefAttrPos[iPtIdx][1], RefAttrPos[iPtIdx][2])

[0412] jMorton = Morton(RefAttrPos[jPtIdx][0], RefAttrPos[jPtIdx][1], RefAttrPos[jPtIdx][2])

[0413] if (iMorton > jMorton)

[0414] Swap(ParentLodPtIdx[i], ParentLodPtIdx[j])

[0415] }

[0416] }

[0417] Generation of a single detail level

[0418] A coarser detail level Lvl + 1 is generated by subsampling points of detail level Lvl.

[0419] When fgs_layer_group_enabled is equal to 1, block-based subsampling (10.6.5.8) is performed.

[0420] Block-based subsampling

[0421] Block-based subsampling generates subsampled output detail levels in the following ways.

[0422] The input detail level is spatially partitioned into a lattice of cubic blocks of size 2^{BlkSizeLog2}.

[0423] Blocks are grouped together in Morton order based on the number of points they contain.

[0424] Assign one point from each block group to a subsampled detail level.

[0425] BlkSizeLog2 := lod_initial_dist_log2 + lod_dist_log2_offset + Lvl + 1

[0426] Under certain conditions, blocks correspond to nodes in the occupancy tree. For example, this is the case when lod_scalability_enabled is 1.

[0427] You must generate a list of block groups by traversing the input detail levels in canonical order. Consecutive blocks must be grouped together until the group contains at least minGrpPts points.

[0428] minGrpPts := lod_scalability_enabled ? 0 : 2 + lod_sampling_period_minus2[Lod]

[0429] The array grpBdry with element grpBdry[grpIdx] identifies block group boundaries as indices of the input detail level array InLodPtIdx.

[0430] for (i = 1, grpStart = 0; i < InLodPtCnt; i++) {

[0431] ptIdx = InLodPtIdx[i]

[0432] ptIdxPrev = InLodPtIdx[i - 1]

[0433] bdryS = (AttrPos[ptIdx][0] ^ AttrPos[ptIdxPrev][0]) >> BlkSizeLog2

[0434] bdryT = (AttrPos[ptIdx][1] ^ AttrPos[ptIdxPrev][1]) >> BlkSizeLog2

[0435] bdryV = (AttrPos[ptIdx][2] ^ AttrPos[ptIdxPrev][2]) >> BlkSizeLog2

[0436] if (bdryS | bdryT | bdryV)

[0437] if (i - grpStart ≥ minGrpPts)

[0438] grpBdry[grpCnt++] = grpStart = i

[0439] }

[0440] grpBdry[grpCnt++] = InLodPtCnt

[0441] For each block group, a test (10.6.5.9) must be performed to determine the index of the point to be assigned to the output detail level. All other points are assigned to the refinement list. The variables GrpStart and GrpEnd identify the start and end of the block group. The result of the test is the variable IdxOfSubsampledPoint.

[0442] OutLodPtCntSize = outRfmtPtCnt = 0

[0443] for (GrpStart = grpIdx = 0; grpIdx < grpCnt; GrpStart = grpBdry[grpIdx++]) {

[0444] GrpEnd = grpBdry[grpIdx]

[0445] … / * IdxOfSubsampledPoint = Point-by-point test result (10.6.5.9) * /

[0446] for (i = GrpStart; i < GrpEnd; i++) {

[0447] if (IdxOfSubsampledPoint == i)

[0448] OutLodPtIdx[OutLodPtCnt++] = InLodPtIdx[i]

[0449] else

[0450] OutRfmtPtIdx[outRfmtPtCnt++] = InLodPtIdx[i]

[0451] }

[0452] }

[0453] Per block-group decision for block-based subsampling

[0454] The derivation of IdxOfSubsampledPoint specifies the input detail level index of the point within the block group to be assigned to the output detail level.

[0455] To select the point to be assigned to the output detail level, the distance to the block group centroid must be used. The block group centroid and point distances must be calculated using attribute coordinates quantized by Exp2(BlkSizeLog2 - 1). The distance metric used is the Manhattan distance.

[0456] FIG. 23 shows block group subsampling (minGrpPts = 3) according to the embodiments.

[0457] The method / device according to the embodiments may include and perform the block group subsampling method according to FIG. 23.

[0458] Figure 23 shows an example of a point-by-point determination. Subsampling creates two detail levels. To subsample LoD_0, points are grouped into block groups containing at least three points. In this example, the block size for LoD_0 (Lvl = 0) is BlkSizeLog2 = 1. The first block group (solid shading) consists of a total of four points from three blocks, each containing one, one, and two points, respectively. The first point closest to the center of the points within the block group is assigned to LoD_1.

[0459] The block group center is the value obtained by dividing centroidSum, which is the sum of all quantized attribute coordinates, by numPtsInGrp, which is the number of points in the block group.

[0460] numPtsInGrp := GrpEnd - GrpStart

[0461] for (k = 0; k < 3; k++)

[0462] centroidSum[k] = 0

[0463] for (i = 0; i < numPtsInGrp; i++) {

[0464] ptIdx = InLodPtIdx[GrpStart + i]

[0465] for (k = 0; k < 3; k++)

[0466] centroidSum[k] += AttrPos[ptIdx][k] >> BlkSizeLog2 - 1

[0467] }

[0468] The array ptDist maps the index of each point within the block group to the distance between that point and the center.

[0469] for (i = 0; i < numPtsInGrp; i++) {

[0470] ptIdx = InLodPtIdx[GrpStart + i]

[0471] ptDist[i] = 0

[0472] for (k = 0; k < 3; k++) {

[0473] posk = AttrPos[ptIdx][k] >> BlkSizeLog2 - 1

[0474] ptDist[i] += Abs(posk × numPtsInGrp - centroidSum[k])

[0475] }

[0476] }

[0477] Assign the point closest to the center of the block group to the output detail level. If the block group contains multiple closest points, the selected point is the closest point that satisfies the following conditions.

[0478] If lod_scalability_enabled is 1 and Lvl is odd: the largest InLodPtIdx index.

[0479] If lod_scalability_enabled is 0 or Lvl is even: the smallest InLodPtIdx index.

[0480] last := lod_scalability_enabled Lvl & 1 : 1

[0481] minIdx = 0

[0482] for (i = 1; i < numPtsInGrp; i++)

[0483] if (last ? dist[i] ≤ dist[minIdx] : dist[i] < dist[minIdx])

[0484] minIdx = i

[0485] IdxOfSubsampledPoint = GrpStart + minIdx

[0486] proposal

[0487] Block-based subsampling

[0488] To complete the generation of detail levels in FGS attribute decoding, we propose specifying a block-based subsampling method such that the partial level of detail (partial LoD) structure corresponds to the partial occupancy tree of FGS.

[0489] Block-based subsampling

[0490] Block-based subsampling generates subsampled output detail levels in the following ways.

[0491] The input detail level is spatially partitioned into a cubic block grid of size 2^{BlkSizeLog2}.

[0492] Group the blocks together in a specific order based on the number of points they contain.

[0493] Assign one point from each block group to a subsampled detail level.

[0494] BlkSizeLog2 := lod_initial_dist_log2 + lod_dist_log2_offset + Lvl + 1

[0495] When fgs_layer_group_enabled is 1, blocks correspond to nodes in the occupancy tree. (lod_initial_dist_log2 and lod_dist_log2_offset can be 0.)

[0496] Generate a block group list by traversing the input detail levels in standard order. Consecutive blocks must be grouped together until the group contains the minimum number of points (minGrpPts).

[0497] minGrpPts := fgs_layer_group_enabled ? 0 : 2 + lod_sampling_period_minus2[Lod]

[0498] If the FGS layer group enable information (fgs_layer_group_enabled) is 1, the minimum point count (minGrpPts) can be set to 0. This may mean that in FGS mode, blocks correspond to nodes in the occupancy tree and are processed regardless of the point count. And if the FGS layer group enable information (fgs_layer_group_enabled) is 0, the minimum point count (minGrpPts) can be set to the value obtained by adding 2 to the LoD sampling period information (lod_sampling_period_minus2[Lod]).

[0499] Block group determination for block-based subsampling

[0500] The point closest to the block group center point must be assigned to the output detail level. If the block group contains multiple closest points, the selected point is the closest point that satisfies the following conditions.

[0501] If fgs_layer_group_enabled is 0 or Lvl is odd: the largest InLodPtIdx index.

[0502] If fgs_layer_group_enabled is 1 and Lvl is even: the smallest InLodPtIdx index.

[0503] last := fgs_layer_group_enabled ? Lvl & 1 : 1

[0504] minIdx = 0

[0505] for (i = 1; i < numPtsInGrp; i++)

[0506] if (last ? ptDist[i] ≤ ptDist[minIdx] : ptDist[i] < ptDist[minIdx])

[0507] minIdx = i

[0508] IdxOfSubsampledPoint = GrpStart + minIdx

[0509] Additionally, we propose adding a condition to the attribute parameter set to prevent unnecessary signals used in non-block-based subsampling methods.

[0510] if last := fgs_layer_group_enabled ? Lvl & 1 : 1 / fgs_layer_group_enabled = 1, use the value of Lvl & 1, otherwise assign 1 to the last variable.

[0511] minIdx = 0 / Initialize minIdx to 0.

[0512] for (i = 1; i < numPtsInGrp; i++) / Iterates by incrementing index i from 1 one by one until numPtsInGrp, which is the number of points in the group.

[0513] if (last ? ptDist[i] ≤ ptDist[minIdx] : ptDist[i] < ptDist[minIdx]) / if last is 1, check if the condition ptDist[i] ≤ ptDist[minIdx] (less than) is true, and if last is 0, check if the condition ptDist[i] < ptDist[minIdx] (less than) is true. Depending on whether the last value is 1 or 0, it can be determined whether to maintain the existing value or update it to a new value when the distance to point index I is equal to the minimum distance.

[0514] minIdx = i / If the above condition is true, update the current index i to the new minIdx (minimum index) and derive it.

[0515] IdxOfSubsampledPoint = GrpStart + minIdx / Determine IdxOfSubsampledPoint (index of the subsampled point) by adding minIdx to GrpStart, the starting position of the group.

[0516] FIG. 24 shows the attribute parameter set data unit syntax according to the embodiments.

[0517] The method / device according to the embodiments can generate a bitstream including the syntax structure of FIG. 24, or acquire a bitstream and decode geometry data or attribute data within the bitstream based on the acquired information.

[0518] LoD scalability_enabled specifies whether attribute values ​​are encoded using restricted LoD generation and predictor search (enabled if 1, disabled if 0). When equal to 1, attribute values ​​can be reconstructed for partially decoded occupancy trees as specified in Annex D. When lod_scalability_enabled is not present, it is inferred to be 0.

[0519] The requirement for bitstream suitability is that lod_scalability_enabled must be 0 if any of the following conditions are true.

[0520] If geom_tree_type is 1, or

[0521] If occtree_coded_axis_list_present is 1, or

[0522] If geom_scaling_enabled is 1 and geom_qp_mul_log2 is not 3, or

[0523] When pred_blending_enabled is 1.

[0524] If pred_max_range_minus1 plus 1 exists, it specifies the distance threshold at which point predictor candidates should be discarded during predictor set pruning for scalable attribute coding. The distance is specified in units of block sizes per detail level.

[0525] The value obtained by adding 1 to the maximum number of LoD levels (lod_max_levels_minus1) specifies the maximum number of detail levels that can be generated by the LoD generation process. When lod_max_levels_minus1 does not exist, it is inferred as MaxSliceDimLog2 - 1.

[0526] The attribute canonical_order_enabled specifies whether the order in which point attributes are encoded is the canonical order in which points are output by the geometry decoding process specified in this document (1 for canonical order, 0 for non-canonical order). If attr_canonical_order_enabled does not exist, it is inferred to be 0.

[0527] The LoD decimation mode (lod_decimation_mode) specifies the decimation method used to generate detail levels. Valid values ​​are specified in Fig. 24. Other values ​​are reserved for future use by ISO / IEC. Decoders compliant with this version of the document must ignore (remove and discard from the bitstream) attribute data units encoded with the reserved value of lod_decimation_mode.

[0528] FIG. 25 shows the interpretation according to the value of lod_decimation_mode according to the embodiments.

[0529] Referring to FIG. 25, lod_decimation_mode is 0, indicating no decimation. lod_decimation_mode is 1, indicating periodic subsampling. lod_decimation_mode is 2, indicating block-based subsampling.

[0530] The value of minus 2 (lod_sampling_period_minus2[ lvl ]) plus 2 specifies the sampling period used to sample points at the detail level lvl to generate the next coarser detail level lvl + 1 when creating the LoD.

[0531] LoD initial distance log2 (lod_initial_dist_log2) specifies the block size at the finest detail level used for LoD generation and predictor search. When lod_initial_dist_log2 does not exist, it is inferred to be 0.

[0532] The LoD distance log2 offset existence information (lod_dist_log2_offset_present) specifies whether the per-slice block size offset specified by lod_dist_log2_offset exists in the ADU header (exists if 1, does not exist if 0). When lod_dist_log2_offset_present does not exist, it is inferred to be 0.

[0533] Referring to FIGS. 24 and 25, the attribute parameter set (attribute_parameter_set()) syntax structure may include a syntax element that determines the LOD configuration method of attribute data in point cloud compression.

[0534] lod_scalability_enabled indicates with 1 bit whether to use the scalability feature. If LOD scalability is enabled (lod_scalability_enabled = 1), parse pred_max_range_minus1. If LOD scalability is not enabled (lod_scalability_enabled = 0) and FGS layer groups are disabled (fgs_layer_group_enabled = 0), parse lod_max_levels_minus1.

[0535] If there is only one LOD level (lod_max_levels_minus1 = 0), parse attr_canonical_order_enabled. If there are multiple LOD levels (2 or more) (lod_max_levels_minus1 > 0), parse the decimation mode for LOD generation (lod_decimation_mode).

[0536] If the LOD decimation mode is greater than 0 (lod_decimation_mode > 0), that is, if a specific subsampling method is defined according to the table in Fig. 25, the following loop is executed. It iterates for the maximum number of levels to configure each LOD level using the for(lvl = 0; lvl < lod_max_levels_minus1; lvl++) statement, and parses the value (lod_sampling_period_minus2[lvl]) that determines the subsampling period at that level. Then, the log value (lod_initial_dist_log2) representing the initial distance of the first LOD is read, and information on whether an offset value for distance calculation exists (lod_dist_log2_offset_present) is parsed.

[0537] FGS Attribute Coding

[0538] Attribute encoding process

[0539] FIG. 26 illustrates an FGS encoding method according to embodiments.

[0540] The encoding method and apparatus according to the embodiments (transmitting device (10000) in FIG. 1, acquisition (20000) to transmission (20002) in FIG. 2, encoder in FIG. 3, encoder in FIG. 8, each device in FIG. 10, encoder in FIG. 21, bitstream and parameter generation in FIG. 18 to 20, bitstream and parameter generation in FIG. 24 to 25, encoding in FIG. 26, method in FIG. 28) may include and perform the method in FIG. 26.

[0541] FIG. 26 is an example flowchart of a layer-group slicing encoder for LoD-based attribute coding.

[0542] First, parameters such as SPS, GPS, APS, and LGSI are generated, and a layer-group structure is configured. The configuration of the layer-group structure can be performed during or before LGSI generation. After geometry coding is performed, LoD generation is carried out based on the geometry nodes. After matching attribute coding layers (e.g., LoD, RAHT coding layers) with layer groups, quantization weights are calculated based on the LoD. Subsequently, for each attribute coding layer, a check is performed to determine whether the layer group has changed, and for the nodes within the attribute coding layer, a check is performed to determine whether the subgroup has changed. If a subgroup or layer group changes, the attribute encoder for the previous subgroup is saved, and the attribute encoder for the current subgroup is used; this ensures continuity of context and zerorun within the subgroup, as well as independence from neighboring subgroups. Once compression of the nodes belonging to all attribute coding layers is complete, the coded bitstream for each subgroup is packaged into a slice. Although the above explanation is based on LoD-based attribute coding, the content covered in this document can be applied to other attribute coding methods such as RAHT.

[0543] Figure 26 shows a flowchart of a layer group slicing encoder. First, the SPS, GPS, APS, and LGSI generation steps generate sequence (SPS), geometry (GPS), attribute (APS), and layer group structure information (LGSI) parameters. This is the process of drawing the blueprint for the entire encoding. Then, the geometry encoder step first encodes the location information (geometry) of points that serve as the basis for attribute coding.

[0544] Subsequently, the LoD generation step generates Levels of Detail (LoD) based on the encoded geometry nodes. Then, the LoD-to-layer-group mapping step matches each generated LoD layer to a predefined layer group. This is a unit partitioning step for parallel processing or independent access. The weight derivation step calculates weights to be used for quantization based on the characteristics of each LoD.

[0545] Subsequently, the Determine layer-group and Determine subgroup steps determine the layer group currently to be processed and the subgroups within it.

[0546] Subsequently, it determines whether the subgroup has changed (Subgroup changed?). If the subgroup has changed (Yes), the Save and load encoder process is performed. The encoder state (context, etc.) of the previous subgroup is saved, and the state of the new subgroup is loaded. This ensures data independence between subgroups while maintaining continuity within the sliced ​​unit. If the subgroups remain the same (No), the existing state is maintained, and node attribute encoding is performed. A loop running from bottom to top is repeated for all nodes and all LoD layers until processing is complete. Then, in the Generate Attribute Data Unit Headers step, data unit headers necessary to package the encoded data into slices are generated. Finally, in the Generate Attribute Bitstreams step, the encoded bitstreams contained in each slice are generated, completing the encoding process.

[0547] Attribute decoder process

[0548] FIG. 27 illustrates an FGS decoding method according to embodiments.

[0549] The decoding method and apparatus according to the embodiments (receiving device (10004) of FIG. 1, receiving (20002) to rendering (20004) of FIG. 2, decoder of FIG. 7 and FIG. 9, each device of FIG. 10, decoder of FIG. 22, bitstream and parameter acquisition of FIG. 18 to FIG. 20, bitstream and parameter acquisition of FIG. 24 to FIG. 25, decoding of FIG. 27, method of FIG. 29) may include and perform the method of FIG. 27.

[0550] The inputs to the attribute decoder are the attribute fine granularity slice bitstream, decoded geometry data, and layer-group structure. Based on the decoded geometry data provided after geometry decoding, Levels of Detail (LoD) generation is performed, and weight derivation is performed based on the relationships between nodes according to the LoD. The information generated at this stage is used in subsequent attribute decoding processes. The first attribute slice can operate identically to the previous attribute slice. Subsequently, if layer-group slicing is enabled, the dependent attribute data unit header is parsed, and layer_group_id and subgroup_id information can be obtained. If necessary, information regarding the LoD can be signaled. Based on this, a layer group matching the current LoD can be found, and a parent subgroup can be selected. When the layer group structure is applied identically to geometry and attributes, as in this document, the reference and parent information used in geometry decoding can be utilized. Based on this, information from dependent attribute data units can be decoded, and point cloud data can be reconstructed.

[0551] The LoD generation and weight derivation methods used in the encoder can be utilized for decoding. Therefore, the process of mapping LoDs to layer groups is explained further below.

[0552] Figure 27 shows a flowchart of the FGS decoding method. First, the SPS, GPS, APS, and LGSI parsing steps prepare decoding settings by parsing sequence, geometry, attribute, and layer group structure information from the bitstream. Then, the geometry decoder step decodes the geometry bitstream to restore location information of the points. The LOD generation step generates a Level of Detail (LoD) based on the restored geometry data in the same way as the encoder. Furthermore, the weight derivation step derives weights necessary for quantization inversion, etc., based on the relationships between nodes within the LoD. This information is subsequently used to restore attribute values ​​to actual colors / reflectance, etc. The attribute data unit header parsing step and the attribute data unit decoding step parse the header of the first or primary attribute data unit and decode the actual data.

[0553] Then, determine whether the layer-group slicing function is enabled (Layer-group slicing enabled?). If it is not enabled (No), proceed to the next step without additional layer group processing. If it is enabled (Yes), perform the dependent data decoding procedure below.

[0554] In the dependent attribute data unit header parsing step, the header of the dependent attribute data is parsed to obtain the layer group ID (layer_group_id) and subgroup ID (subgroup_id). Then, in the LoD to layer-group mapping step, the layer group to which the currently decoding LoD belongs is matched based on the parsed information. Subsequently, in the parent subgroup selection step, the parent subgroup that the current subgroup should reference is selected. At this stage, if the geometry and attribute structures are the same, the reference information used during geometry restoration can be utilized as is. Finally, in the dependent attribute data unit decoding step, the current dependent attribute data is finally decoded by referencing the parent subgroup information. Afterward, the loop is repeated until all bitstreams are processed. Then, the restored geometry and attribute data are combined to reconstruct the final point cloud (Generate output point cloud).

[0555] FIG. 28 illustrates a encoding method according to embodiments.

[0556] The encoding method of Fig. 28 can follow the inverse process of the decoding method of Fig. 29.

[0557] The method according to the embodiments may include the step of encoding geometry data of point cloud data (S2810) / or the step of encoding attribute data of point cloud data (S2820).

[0558] The encoding method and apparatus according to the embodiments may include and perform a transmitting device (10000) of FIG. 1, an acquisition (20000) to a transmission (20002) of FIG. 2, an encoder of FIG. 3 and FIG. 8, each device of FIG. 10, an encoder of FIG. 21, bitstream and parameter generation of FIG. 18 to 20, FIG. 24 to 25, encoding of FIG. 26, and a method of FIG. 28.

[0559] Referring to the general generation process, the step of encoding attribute data according to the embodiments (S2820) may include the step of generating detail levels based on the minimum detail level and maximum detail level derived from the geometry data of the FGS (Fine Granularity Slice) based on the fact that the FGS layer group activation information (fgs_layer_group_enabled) is 1.

[0560] Referring to the general generation process, the step of generating detail levels according to the embodiments may include the step of generating an output detail level (Lvl+1) by block-based subsampling points of a detail level (Lvl) based on the fact that the FGS layer group activation information (fgs_layer_group_enabled) is 1.

[0561] Referring to the general generation process, the step of generating an output detail level (Lvl+1) according to the embodiments can be performed repeatedly starting from the minimum detail level until only a single point remains or the maximum detail level is reached.

[0562] The encoding method of FIG. 28 includes a memory and at least one processor connected to the memory, and the encoding device may be configured to encode geometry data of point cloud data and attribute data of point cloud data.

[0563] The device according to the embodiments may include a computer-readable storage medium that stores a bitstream generated by the encoding method of FIG. 28.

[0564] A method according to the embodiments may include a method comprising the steps of: acquiring a bitstream for point cloud data; generating the bitstream based on the steps of encoding geometry data of the point cloud data and encoding attribute data of the point cloud data; and transmitting data including the bitstream.

[0565] FIG. 29 illustrates a decoding method according to embodiments.

[0566] The decoding method of Fig. 29 can follow the inverse process of the encoding method of Fig. 28.

[0567] The method according to the embodiments may include the step of decoding geometry data of point cloud data within a bitstream (S2910); and / or the step of decoding attribute data of point cloud data (S2920); etc.

[0568] The decoding method and apparatus according to the embodiments may include and perform a receiving device (10004) of FIG. 1, a receiving device (20002) to a rendering device (20004) of FIG. 2, a decoder of FIG. 7 and FIG. 9, each device of FIG. 10, a decoder of FIG. 22, bitstream and parameter acquisition of FIG. 18 to FIG. 20, FIG. 24 to FIG. 25, decoding of FIG. 27, a method of FIG. 29, etc.

[0569] Referring to the general generation process, the step (S2920) of decoding attribute data of point cloud data according to the embodiments may include the step of generating detail levels based on the minimum detail level and maximum detail level derived from the geometry data of the FGS (Fine Granularity Slice) based on the fact that the FGS layer group activation information (fgs_layer_group_enabled) is 1.

[0570] Referring to the general generation process, the step of generating detail levels according to the embodiments may include the step of generating an output detail level (Lvl+1) by block-based subsampling points of a detail level (Lvl) based on the fact that the FGS layer group activation information (fgs_layer_group_enabled) is 1.

[0571] Referring to the general generation process, the step of generating an output detail level (Lvl+1) according to the embodiments can be performed repeatedly starting from the minimum detail level until only a single point remains or the maximum detail level is reached.

[0572] Referring to the general generation process, the maximum detail level according to the embodiments overlaps with the minimum detail level of the parent FGS of the FGS, and the maximum detail level is derived based on the logarithm of the node size of the FGS geometry based on the layer group identifier to which the FGS belongs being 0, and based on the value obtained by adding 1 to the logarithm of the node size of the FGS geometry based on the layer group identifier to which the FGS belongs being greater than 0. In this case, the node size may refer to the node size of the root node of the FGS or the node size of the coarsest level of the FGS.

[0573] Referring to block-based subsampling, block-based subsampling according to the embodiments includes the step of spatially partitioning the detail level based on blocks, and based on FGS layer group activation information (fgs_layer_group_enabled) being 1, the blocks may be associated with nodes of the occupancy tree of the point cloud data.

[0574] Referring to block-based subsampling, block-based subsampling according to the embodiments may further include the step of grouping blocks until the block group includes at least a minimum number of points.

[0575] Referring to block-based subsampling, the minimum number of points according to the embodiments can be derived to 0 based on the FGS layer group activation information (fgs_layer_group_enabled) being 1. And based on the FGS layer group activation information (fgs_layer_group_enabled) being 0, it can be derived based on the detail level sampling period information.

[0576] Referring to the determination per block group for block-based subsampling, block-based subsampling according to the embodiments may further include the step of selecting the point closest to the center point of the block group from the block group, and the step of assigning the selected point to an output detail level.

[0577] Referring to the block group decision for block-based subsampling, when there are multiple points closest to the center point, the step of selecting the point closest to the center point of the block group may further include the step of selecting the point with the largest index based on the fact that the FGS layer group activation information (fgs_layer_group_enabled) is 0 or the detail level index is odd, and the step of selecting the point with the smallest index based on the fact that the FGS layer group activation information (fgs_layer_group_enabled) is 1 and the detail level index is even.

[0578] The bitstream according to the embodiments may include FGS layer group enable information (fgs_layer_group_enabled) indicating that the slice contains the FGS of the partial slice attribute. According to the embodiments, the FGS layer group enable information (fgs_layer_group_enabled) may be included in the Sequence parameter set (SPS) syntax. Referring to FIG. 24, when the FGS layer group enable information (fgs_layer_group_enabled) has a value of 0, i.e., when the FGS layer group is disabled, the LOD maximum level information (lod_max_levels_minus1) and subsequent LoD syntax elements may be parsed.

[0579] The decoding method of FIG. 29 includes a memory and at least one processor connected to the memory, and the decoding device may be configured to decode geometry data of point cloud data and attribute data of point cloud data.

[0580] The effects according to the features of the present invention are described from the perspective of the transmitting end.

[0581] When compressing point cloud data based on the present invention, it is expected that attribute information can be compressed at the transmitting end with a small amount of computation. Therefore, it is expected to be usable in transmission systems requiring low-delay.

[0582] Furthermore, when performing point cloud compression based on the point cloud data structure proposed in this invention, it is expected that receivers of various performance levels can be supported based on a single compressed bitstream depending on the compression method. For example, when compressing information for decoders of various performance levels, since receivers of various performance levels can be supported through a single bitstream instead of generating or storing independent compression information tailored to each decoder's performance, it is expected to offer advantages in terms of transmitter storage space and bit efficiency. Additionally, in cases where transmission bandwidth is limited, there is an advantage in that low-resolution point cloud data can be generated and transmitted from the transmitter.

[0583] The effects according to the features of the present invention are described from the perspective of the receiving end.

[0584] When receiving point cloud data based on the present invention, the receiving unit recovers attribute information almost simultaneously with geometry decoding with a small amount of computation, so it is expected to be usable in a transmission and reception system where low-delay is required.

[0585] In addition, when selecting the output level of attribute information, it is expected that attribute information suitable for the performance of the receiver can be output without delay, even by receivers with low computational power. For example, different results may be output during attribute decoding and reconstruction depending on the performance of the receiver or the requirements of the system. In this case, the attributes of each level decoded or reconstructed can be used as attribute values ​​matched with the octree node of that level. When a colorized octree is transmitted based on point cloud coding characteristics, the octree level can be selected according to output performance or renderer performance. Alternatively, considering the output or rendering performance of the receiver, octree colorization can be performed on the point cloud data restored after decoding, and then a low-resolution image can be output or rendered.

[0586] When a bitstream is divided into slices and transmitted according to the method proposed in this document, the receiver can selectively transmit the bitstream to the decoder based on the density of the point cloud data to be represented, depending on the decoder performance or application field. In this case, since the selection is made before decoding, the decoder efficiency is increased, and there is an advantage in being able to support decoders of various performance levels.

[0587] The embodiments have been described in terms of methods and / or devices, and the description of the methods and the description of the devices may be applied complementarily.

[0588] Although the drawings have been described separately for the convenience of explanation, it is also possible to design a new embodiment by combining the embodiments described in each drawing. Furthermore, designing a computer-readable recording medium containing a program for executing the previously described embodiments, as required by a person skilled in the art, falls within the scope of the embodiments. The apparatus and method according to the embodiments are not limited to the configuration and method of the embodiments described above; rather, the embodiments may be configured by selectively combining all or part of each embodiment to allow for various modifications. Although preferred embodiments have been illustrated and described, the embodiments are not limited to the specific embodiments described above. It is not only possible for a person skilled in the art to make various modifications without departing from the essence of the embodiments claimed in the claims, but such modifications should not be understood individually from the technical concept or perspective of the embodiments.

[0589] Various components of the device of the embodiments may be implemented by hardware, software, firmware, or a combination thereof. Various components of the embodiments may be implemented as a single chip, for example, a single hardware circuit. Depending on the embodiments, the components according to the embodiments may each be implemented as separate chips. Depending on the embodiments, at least one of the components of the device according to the embodiments may be composed of one or more processors capable of executing one or more programs, and one or more programs may include instructions for performing or executing any one or more of the operations / methods according to the embodiments. Executable instructions for performing the methods / operations of the device according to the embodiments may be stored in non-transient CRMs or other computer program products configured to be executed by one or more processors, or may be stored in transient CRMs or other computer program products configured to be executed by one or more processors. Additionally, memory according to the embodiments may be used as a concept that includes not only volatile memory (e.g., RAM, etc.) but also non-volatile memory, flash memory, PROM, etc. In addition, it may also include implementation in the form of carrier waves, such as transmission over the Internet. Furthermore, processor-readable recording media are distributed across networked computer systems, allowing processor-readable code to be stored and executed in a distributed manner.

[0590] In this document, “ / ” and “,” are interpreted as “and / or.” For example, “A / B” is interpreted as “A and / or B,” and “A, B” is interpreted as “A and / or B.” Additionally, “A / B / C” means “at least one of A, B and / or C.” Also, “A, B, C” means “at least one of A, B and / or C.” Additionally, in this document, “or” is interpreted as “and / or.” For example, “A or B” may mean 1) “A” alone, 2) “B” alone, or 3) “A and B.” In other words, “or” in this document may mean “additionally or alternatively.”

[0591] Terms such as "first," "second," etc., may be used to describe various components of the embodiments. However, the interpretation of the various components according to the embodiments should not be limited by these terms. These terms are merely used to distinguish one component from another. For example, the first user input signal may be referred to as the second user input signal. Similarly, the second user input signal may be referred to as the first user input signal. The use of these terms should be interpreted as not departing from the scope of the various embodiments. Although the first user input signal and the second user input signal are both user input signals, they do not imply the same user input signals unless clearly indicated in the context.

[0592] The terms used to describe the embodiments are intended for the purpose of describing specific embodiments and are not intended to limit the embodiments. As used in the description of the embodiments and in the claims, the singular is intended to include the plural unless explicitly indicated in the context. Expressions of and / or are used to mean including all possible combinations between the terms. Expressions of include describe the presence of features, numbers, steps, elements, and / or components and do not imply the exclusion of additional features, numbers, steps, elements, and / or components. Conditional expressions such as "if" or "when" used to describe the embodiments are not limited to being optional. It is intended to be interpreted as "when a specific condition is satisfied," "when a related action is performed in response to a specific condition," or "when a related definition is interpreted."

[0593] Additionally, operations according to the embodiments described herein may be performed by a transmitting and receiving device including memory and / or a processor, depending on the embodiments. The memory may store programs for processing / controlling operations according to the embodiments, and the processor may control various operations described in this document. The processor may be referred to as a controller, etc. Operations in the embodiments may be performed by firmware, software, and / or a combination thereof, and the firmware, software, and / or a combination thereof may be stored in the processor or in memory.

[0594] Meanwhile, the operation according to the embodiments described above may be performed by a transmitting device and / or a receiving device according to the embodiments. The transmitting and receiving device may include a transmitting and receiving unit for transmitting and receiving media data, a memory for storing instructions (program code, algorithm, flowchart and / or data) for a process according to the embodiments, and a processor for controlling the operations of the transmitting and receiving devices.

[0595] The processor may be referred to as a controller, etc., and may correspond, for example, to hardware, software, and / or a combination thereof. The operation according to the embodiments described above may be performed by the processor. Additionally, the processor may be implemented as an encoder / decoder, etc., for the operation of the embodiments described above.

[0596]

[0597] As described above, the relevant details have been explained in the best mode for carrying out the embodiments.

[0598]

[0599] As described above, the embodiments may be applied wholly or partially to point cloud data transmission and reception devices and systems.

[0600] Those skilled in the art may make various changes or modifications to the embodiments within the scope of the embodiments.

[0601] The embodiments may include modifications / variations, and such modifications / variations do not exceed the scope of the claims and their equivalents.

Claims

1. A step of decoding geometry data of point cloud data within a bitstream; and A step of decoding attribute data of the above point cloud data; comprising method.

2. In Paragraph 1, The step of decoding the above attribute data is, A step comprising generating detail levels based on the minimum detail level and maximum detail level derived from the geometry data of the Fine Granularity Slice (FGS), based on the fact that the FGS layer group activation information (fgs_layer_group_enabled) is 1, method.

3. In Paragraph 2, The step of generating the above detail levels is, Based on the fact that the above FGS layer group activation information (fgs_layer_group_enabled) is 1, the method includes the step of generating an output detail level (Lvl+1) by block-based subsampling points of the detail level (Lvl), and The step of generating the output detail level (Lvl+1) is performed repeatedly starting from the minimum detail level until only a single point remains or the maximum detail level is reached. method.

4. In Paragraph 3, The above maximum detail level overlaps with the minimum detail level of the parent FGS of the above FGS, and The above maximum detail level is: Based on the fact that the layer group identifier to which the above FGS belongs is 0, it is derived based on the logarithm of the node size of the above FGS geometry, and Based on the layer group identifier to which the above FGS belongs being greater than 0, derived based on the value obtained by adding 1 to the logarithm of the node size of the above FGS geometry, method.

5. In Paragraph 4, The above block-based subsampling includes the step of dividing the space based on blocks at the detail level, and Based on the fact that the above FGS layer group activation information (fgs_layer_group_enabled) is 1, the block is related to a node of the occupancy tree of the point cloud data, method.

6. In Paragraph 5, The above block-based subsampling further includes the step of grouping the blocks until the block group contains points greater than or equal to a minimum number of points, and The minimum number of points mentioned above is: Based on the fact that the above FGS layer group activation information (fgs_layer_group_enabled) is 1, it is induced to 0, and Based on the fact that the above FGS layer group activation information (fgs_layer_group_enabled) is 0, derived based on the detail level sampling period information, method.

7. In Paragraph 6, The above block-based subsampling is: A step of selecting the point closest to the center point of the block group from the block group; and The method further includes the step of assigning the selected point to the output detail level, If there are multiple points closest to the center point, the step of selecting the point closest to the center point of the block group is: A step of selecting the point with the largest index based on the fact that the FGS layer group activation information (fgs_layer_group_enabled) is 0 or the index of the detail level is odd; and The step of selecting the point with the smallest index based on the fact that the above FGS layer group activation information (fgs_layer_group_enabled) is 1 and the index of the detail level is even; further comprising method.

8. In Paragraph 1, The above bitstream includes FGS layer group enable information (fgs_layer_group_enabled) indicating that the slice includes the FGS of the partial slice attribute, method.

9. Memory; and At least one processor connected to the memory; comprising, wherein the at least one processor: Decoding geometry data of point cloud data within a bitstream; and Decoding attribute data of the above point cloud data; configured to do so, device.

10. A step of encoding the geometry data of the point cloud data; and A step of encoding attribute data of the above point cloud data; comprising method.

11. In Paragraph 10, The step of encoding the above attribute data is, A method comprising the step of generating detail levels based on a minimum detail level and a maximum detail level derived from geometry data of a Fine Granularity Slice (FGS), method.

12. In Paragraph 11, The step of generating the above detail levels is, It includes a step of generating an output detail level (Lvl+1) by block-based subsampling points of the detail level (Lvl), and The step of generating the above output detail level is performed repeatedly, starting from the above minimum detail level until only a single point remains or until the above maximum detail level is reached. method.

13. Memory; and At least one processor connected to the memory; comprising, wherein the at least one processor: Encoding the geometry data of the point cloud data; and Configured to encode the attribute data of the above point cloud data; device.

14. A computer-readable storage medium for storing a bitstream generated by the method according to paragraph 10.

15. Step for acquiring a bitstream for point cloud data, The bitstream is generated based on the step of encoding geometry data of the point cloud data; and the step of encoding attribute data of the point cloud data; and A method comprising the step of transmitting data including the bitstream above.

Images