Global Motion Estimation Using Road and Ground Object Labels for Geometry-Based Point Cloud Compression

JP2024503972A5Pending Publication Date: 2024-12-09QUALCOMM INC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
JP2023532399
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Priority Date
2021-12-21
Filing Date
2021-12-22
Publication Date
2024-12-09

AI Technical Summary

Technical Problem

Existing point cloud encoding and decoding technologies are inefficient in terms of signaling overhead and processing operations, particularly when dealing with stationary ground/road points and moving object points in dynamic environments.

Method used

The proposed technique classifies points in a point cloud as either ground/road points or object points and applies separate encoding/decoding strategies, using global motion vectors for object points and different mechanisms for ground/road points, reducing the need for local motion vectors and improving encoding/decoding efficiency.

Benefits of technology

This approach significantly reduces signaling overhead and processing operations required for point cloud compression, enhancing the efficiency of encoding and decoding devices.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

An exemplary device for coding point cloud data includes a memory configured to store data representing points of the point cloud, and one or more processors implemented in a circuit and configured to determine elevation values ​​of points in the point cloud, classify the points into a set of ground points or a set of object points according to the elevation values, and code the ground points and the object points according to the classification. The one or more processors may determine a top threshold and a bottom threshold, and classify the ground points and the object points according to the top threshold and the bottom threshold. The one or more processors may further code a data structure, such as a geometry parameter set (GPS), that includes data representing the top threshold and the bottom threshold.
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] This application claims priority to U.S. Application No. 17 / 558,362, filed December 21, 2021, and U.S. Provisional Application No. 63 / 131,637, filed December 29, 2020, and U.S. Provisional Application No. 63 / 171,945, filed April 7, 2021, the entire contents of each of which are incorporated herein by reference. U.S. Application No. 17 / 558,362 claims the benefit of U.S. Application No. 63 / 131,637, filed December 29, 2020, and U.S. Provisional Application No. 63 / 171,945, filed April 7, 2021.

[0002] This disclosure relates to point cloud encoding and decoding. [Background technology]

[0003] A point cloud includes a collection of points in 3D space. Each of the points may have a collection of attributes associated with it. The attributes may be color information, such as R, G, B or Y, Cb, Cr information, or reflectance information, or other attributes. Point clouds may be captured by various cameras or sensors, such as LIDAR sensors and 3D scanners. Point clouds may be computer generated. Point cloud data may be used in a variety of applications, including but not limited to construction (modeling), graphics (3D models for visualization and animation), and the automotive industry (LIDAR sensors used to aid navigation).

[0004] A point cloud encoder / decoder (codec) may enclose the 3D space occupied by the point cloud data in a virtual bounding box. The location of the point cloud in the bounding box may be represented with some precision. Thus, the point cloud codec may quantize the location of one or more points based on that precision. At the smallest level, the point cloud codec divides the bounding box into voxels, which are the smallest units of space, represented by a unit cube. A voxel in a bounding box may be associated with zero, one, or more points. The point cloud codec may divide the bounding box into multiple cubic / rectangular regions, sometimes called tiles. The point cloud codec may code the tiles into one or more slices. The partitioning of the bounding box into slices and tiles may be based on the number of points in each partition, or based on other considerations (e.g., a particular region may be coded as a tile). The slice regions may be further partitioned using partitioning decisions similar to those in video codecs. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] U.S. Patent Application Serial No. 17 / 495,428 [Patent Document 2] U.S. Provisional Patent Application No. 63 / 090,657 Summary of the Invention [Means for solving the problem]

[0006] In general, this disclosure describes techniques for coding (encoding and decoding) point cloud data. Specifically, a G-PCC encoder and / or decoder may determine whether a point in a point cloud is a ground / road point or an object point, and then use these classifications of the point to code (encode or decode) the point. For example, the G-PCC encoder or decoder may generate a set of global motion information for only the object points. In some examples, the G-PCC encoder or decoder may further generate a set of global motion information for only the ground / road points. Alternatively, the G-PCC encoder or decoder may code the ground / road points using local motion information and / or intra prediction.

[0007] In one example, a method for coding G-PCC data includes determining elevation values ​​of points in a point cloud, classifying the points into a set of ground points or a set of object points according to the elevation values, and coding the ground points and object points according to the classification.

[0008] In another example, a device for coding G-PCC data includes a memory configured to store data representing points of a point cloud, and one or more processors implemented in circuitry and configured to determine elevation values ​​of the points in the point cloud, classify the points into a set of ground points or a set of object points according to the elevation values, and code the ground points and the object points according to the classification.

[0009] In another example, a computer-readable storage medium storing instructions that, when executed, cause a processor to determine elevation values ​​for points in a point cloud, classify the points into a set of ground points or a set of object points according to the elevation values, and code the ground points and object points according to the classification.

[0010] In another example, a device for coding G-PCC data includes means for determining elevation values ​​of points in a point cloud, means for classifying the points into a set of ground points or a set of object points according to the elevation values, and means for coding the ground points and the object points according to the classification.

[0011] The details of one or more examples are set forth in the accompanying drawings and the description below. Other features, objects, and advantages will become apparent from the description, drawings, and claims. [Brief description of the drawings]

[0012] [Figure 1] FIG. 1 is a block diagram illustrating an example encoding and decoding system that may implement the techniques of this disclosure. [Diagram 2] FIG. 1 is a block diagram illustrating an example geometry point cloud compression (G-PCC) encoder that may be configured to perform the techniques of this disclosure. [Diagram 3] FIG. 1 is a conceptual diagram showing an example of inter-prediction coding in G-PCC. [Figure 4] A block diagram showing an example G-PCC decoder that may be configured to perform the techniques of this disclosure. [Diagram 5] FIG. 1 is a conceptual diagram showing an example of inter-prediction decoding in G-PCC. [Figure 6] FIG. 2 is a conceptual diagram illustrating an example prediction tree that may be used when performing the techniques of this disclosure. [Figure 7] FIG. 2 is a conceptual diagram illustrating an example rotational LIDAR acquisition model. [Figure 8] 1 is a flowchart illustrating an exemplary motion estimation process for the G-PCC InterEM software. [Figure 9] 5 is a flowchart illustrating an example process for estimating global motion. [Figure 10] 5 is a flowchart illustrating an example process for estimating local node motion vectors. [Figure 11]11 is a graph illustrating an example classification of a group into ground (road) and objects using two thresholds of the z-values ​​of points, according to techniques of this disclosure. [Figure 12] 11 is a graph illustrating an example derivation of a threshold using a histogram, in accordance with techniques of this disclosure. [Figure 13] FIG. 1 is a conceptual diagram illustrating labeling points in a cloud as roads and objects according to techniques of this disclosure. [Figure 14] 4 is a flowchart illustrating an example method for encoding a point cloud in accordance with techniques of this disclosure. [Figure 15] 1 is a flowchart illustrating an example method for decoding a point cloud in accordance with techniques of this disclosure. [Figure 16] FIG. 1 is a conceptual diagram illustrating a laser package, such as a LIDAR sensor or other system including one or more lasers, scanning a point in three-dimensional space. [Figure 17] FIG. 9 is a conceptual diagram illustrating an example distance measurement system 900 that may be used with one or more techniques of the present disclosure. [Figure 18] FIG. 1 is a conceptual diagram illustrating an example vehicle-based scenario in which one or more techniques of the present disclosure may be used. [Figure 19] FIG. 1 is a conceptual diagram of an example extended reality system in which one or more techniques of this disclosure may be used. [Figure 20] FIG. 1 is a conceptual diagram illustrating an example mobile device system in which one or more techniques of the present disclosure may be used. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] The point cloud data may be generated, for example, by using a LIDAR system mounted on the automobile. The LIDAR system may emit a laser in multiple different directions in bursts over time as the automobile moves. Thus, for a given laser emission, a point cloud may be formed. To compress the point cloud data, each point cloud (frame) may be coded relative to each other, for example, using intra-frame or inter-frame prediction. This disclosure recognizes that since most objects around the automobile remain relatively stationary, points in the point cloud corresponding to the objects may be predicted using a common global motion vector (which may be expected to roughly correspond to the direction and offset followed by the automobile). However, points on the ground may generally remain static between frames, since the road or ground below the automobile may be expected to be relatively flat, and therefore the laser may be expected to identify points in the same relative location in each frame.

[0014] Therefore, the present disclosure describes techniques that may reduce signaling overhead and coding information. Specifically, a Geometry Point Cloud Compression (G-PCC) encoder and a G-PCC decoder may be configured to separately encode and decode object points and road / ground points. That is, the G-PCC encoder may be configured to classify points in a point cloud as either object points or ground / road points, and then encode the object points using a global motion vector while separately encoding the ground / road points (e.g., using a zero motion vector, a second different global motion vector, respective local motion vectors, intra prediction, or other separate encoding techniques). Similarly, the G-PCC decoder may decode the object points separately from the road / ground points. Using a single global motion vector for all object points in this manner may consume much fewer bits than using respective local motion vectors separately for each of the points in the point cloud. Similarly, coding all road / ground points together may reduce signaling overhead and coding data. These techniques may further reduce the number of processing operations required to encode and decode a point cloud, thus improving the efficiency of operation of G-PCC encoding and decoding devices, as well as the field of geometry-based point cloud compression as a whole.

[0015] 1 is a block diagram illustrating an example encoding and decoding system 100 that may implement the techniques of this disclosure. The techniques of this disclosure are generally directed to coding (encoding and / or decoding) point cloud data, i.e., supporting point cloud compression. In general, point cloud data includes any data for processing a point cloud. Coding may be useful for compressing and / or decompressing the point cloud data.

[0016] As shown in FIG. 1, the system 100 includes a source device 102 and a destination device 116. The source device 102 provides encoded point cloud data to be decoded by the destination device 116. Specifically, in the example of FIG. 1, the source device 102 provides the point cloud data to the destination device 116 via a computer-readable medium 110. The source device 102 and the destination device 116 may comprise any of a wide range of devices, including desktop computers, notebook (i.e., laptop) computers, tablet computers, set-top boxes, telephone handsets such as smartphones, televisions, cameras, display devices, digital media players, video gaming consoles, video streaming devices, terrestrial or marine vehicles, spacecraft, aircraft, robots, LIDAR devices, satellites, and the like. In some cases, the source device 102 and the destination device 116 may be capable of wireless communication.

[0017] In the example of FIG. 1, the source device 102 includes a data source 104, a memory 106, a G-PCC encoder 200, and an output interface 108. The destination device 116 includes an input interface 122, a G-PCC decoder 300, a memory 120, and a data consumer 118. According to this disclosure, the G-PCC encoder 200 of the source device 102 and the G-PCC decoder 300 of the destination device 116 may be configured to apply techniques of this disclosure relating to labeling points in a point cloud as ground points or object points according to height values ​​for the points. Thus, the source device 102 represents an example of an encoding device, and the destination device 116 represents an example of a decoding device. In other examples, the source device 102 and the destination device 116 may include other components or arrangements. For example, the source device 102 may receive data (e.g., point cloud data) from an internal or external source. Similarly, the destination device 116 may interface with external data consumers rather than including the data consumers within the same device.

[0018] The system 100 as shown in FIG. 1 is only an example. In general, other digital encoding and / or decoding devices may perform the techniques of this disclosure relating to labeling points in a point cloud as ground points or object points according to height values ​​for the points. The source device 102 and the destination device 116 are only examples of devices in which the source device 102 generates coded data for transmission to the destination device 116. This disclosure refers to devices that perform coding (encoding and / or decoding) of data as "coding" devices. Thus, the G-PCC encoder 200 and the G-PCC decoder 300 represent examples of coding devices, specifically, encoders and decoders, respectively. In some examples, the source device 102 and the destination device 116 may operate substantially symmetrically such that each of the source device 102 and the destination device 116 includes encoding and decoding components. Thus, the system 100 may support one-way or two-way transmission between the source device 102 and the destination device 116, for example, for streaming, playback, broadcast, telephony, navigation, and other uses.

[0019] In general, the data source 104 represents a source of data (i.e., raw, unencoded point cloud data) and may provide a sequential series of "frames" of data to the G-PCC encoder 200, which encodes the data for the frames. The data source 104 of the source device 102 may include a point cloud capture device, such as any of a variety of cameras or sensors, e.g., a 3D scanner or light detection and ranging (LIDAR) device, one or more video cameras, an archive containing previously captured data, and / or a data feed interface for receiving data from a data content provider. Alternatively or additionally, the point cloud data may be computer-generated from a scanner, camera, sensor, or other data. For example, the data source 104 may generate computer graphics-based data as source data, or produce a combination of live, archived, and computer-generated data. In each case, the G-PCC encoder 200 encodes the captured, pre-captured, or computer-generated data. The G-PCC encoder 200 may reorder the frames from the order in which they were received (sometimes referred to as "display order") into a coding order for coding. The G-PCC encoder 200 may generate one or more bitstreams including the encoded data. The source device 102 may then output the encoded data via the output interface 108 to a computer-readable medium 110, for receipt and / or retrieval by, for example, an input interface 122 of a destination device 116.

[0020] The memory 106 of the source device 102 and the memory 120 of the destination device 116 may represent general purpose memories. In some examples, the memory 106 and the memory 120 may store raw data, e.g., raw data from the data source 104 and raw decoded data from the G-PCC decoder 300. Additionally or alternatively, the memory 106 and the memory 120 may store software instructions executable by, e.g., the G-PCC encoder 200 and the G-PCC decoder 300, respectively. Although the memory 106 and the memory 120 are shown in this example separately from the G-PCC encoder 200 and the G-PCC decoder 300, it should be understood that the G-PCC encoder 200 and the G-PCC decoder 300 may also include internal memories for functionally similar or equivalent purposes. Additionally, the memory 106 and the memory 120 may store encoded data, e.g., the output from the G-PCC encoder 200 and the input to the G-PCC decoder 300. In some examples, portions of memory 106 and memory 120 may be allocated as one or more buffers, e.g., for storing raw decoded and / or encoded data. For example, memory 106 and memory 120 may store data representing a point cloud.

[0021] The computer-readable medium 110 may represent any type of medium or device capable of transporting encoded data from the source device 102 to the destination device 116. In one example, the computer-readable medium 110 represents a communication medium for enabling the source device 102 to transmit encoded data directly to the destination device 116 in real time, for example, via a radio frequency network or a computer-based network. The output interface 108 may modulate a transmission signal including the encoded data, and the input interface 122 may demodulate a received transmission signal, according to a communication standard, such as a wireless communication protocol. The communication medium may comprise any wireless or wired communication medium, such as a radio frequency (RF) spectrum, or one or more physical transmission lines. The communication medium may form part of a packet-based network, such as a local area network, a wide area network, or a global network, such as the Internet. The communication medium may include routers, switches, base stations, or any other equipment that may be useful to facilitate communication from the source device 102 to the destination device 116.

[0022] In some examples, source device 102 may output the encoded data from output interface 108 to storage device 112. Similarly, destination device 116 may access the encoded data from storage device 112 via input interface 122. Storage device 112 may include any of a variety of distributed or locally accessed data storage media, such as a hard drive, a Blu-ray disc, a DVD, a CD-ROM, flash memory, volatile or non-volatile memory, or any other suitable digital storage medium for storing encoded data.

[0023] In some examples, the source device 102 may output the encoded data to a file server 114 or another intermediate storage device, which may store the encoded data generated by the source device 102. The destination device 116 may access the stored data from the file server 114 via streaming or download. The file server 114 may be any type of server device capable of storing the encoded data and transmitting the encoded data to the destination device 116. The file server 114 may represent a web server (e.g., for a website), a file transfer protocol (FTP) server, a content delivery network device, or a network attached storage (NAS) device. The destination device 116 may access the encoded data from the file server 114 through any standard data connection, including an Internet connection. This may include a wireless channel (e.g., a Wi-Fi connection), a wired connection (e.g., a digital subscriber line (DSL), a cable modem, etc.), or a combination of both suitable for accessing the encoded data stored in the file server 114. The file server 114 and the input interface 122 may be configured to operate according to a streaming transmission protocol, a download transmission protocol, or a combination thereof.

[0024] The output interface 108 and the input interface 122 may represent a wireless transmitter / receiver, a modem, a wired networking component (e.g., an Ethernet card), a wireless communication component operating according to any of the various IEEE 802.11 standards, or other physical components. In examples where the output interface 108 and the input interface 122 comprise wireless components, the output interface 108 and the input interface 122 may be configured to transfer data, such as encoded data, according to a cellular communication standard, such as 4G, 4G-LTE (Long-Term Evolution), LTE Advanced, 5G, etc. In some examples where the output interface 108 comprises a wireless transmitter, the output interface 108 and the input interface 122 may be configured to transfer data, such as encoded data, according to other wireless standards, such as the IEEE 802.11 specification, the IEEE 802.15 specification (e.g., ZigBee™), the Bluetooth™ standard, etc. In some examples, the source device 102 and / or the destination device 116 may include respective system-on-chip (SoC) devices. For example, the source device 102 may include a SoC device for performing functions attributed to the G-PCC encoder 200 and / or the output interface 108, and the destination device 116 may include a SoC device for performing functions attributed to the G-PCC decoder 300 and / or the input interface 122.

[0025] The techniques of this disclosure may be applied to encoding and decoding in support of any of a variety of applications, such as communication between autonomous vehicles, communication between scanners, cameras, sensors, and processing devices such as local or remote servers, geographic mapping, or other applications.

[0026] The input interface 122 of the destination device 116 receives the encoded bitstream from the computer-readable medium 110 (e.g., a communication medium, a storage device 112, a file server 114, etc.). The encoded bitstream may include signaling information defined by the G-PCC encoder 200 that is also used by the G-PCC decoder 300, such as syntax elements having values ​​that describe characteristics and / or processing of the coded units (e.g., slices, pictures, groups of pictures, sequences, etc.). The data consumer 118 uses the decoded data. For example, the data consumer 118 may use the decoded data to determine the location of a physical object. In some examples, the data consumer 118 may include a display for presenting an image based on the point cloud.

[0027] The G-PCC encoder 200 and the G-PCC decoder 300 may each be implemented as any of a variety of suitable encoder and / or decoder circuits, such as one or more microprocessors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), discrete logic, software, hardware, firmware, or any combination thereof. When the techniques are implemented partially in software, a device may store instructions for the software on a suitable non-transitory computer-readable medium and execute the instructions in hardware using one or more processors to perform the techniques of this disclosure. Each of the G-PCC encoder 200 and the G-PCC decoder 300 may be included in one or more encoders or decoders, any of which may be integrated as part of a composite encoder / decoder (codec) in the respective device. A device including the G-PCC encoder 200 and / or the G-PCC decoder 300 may comprise one or more integrated circuits, microprocessors, and / or other types of devices.

[0028] The G-PCC encoder 200 and the G-PCC decoder 300 may operate according to a coding standard, such as the Video Point Cloud Compression (V-PCC) standard or the Geometry Point Cloud Compression (G-PCC) standard. This disclosure may generally refer to coding (e.g., encoding and decoding) a picture to include the process of encoding or decoding data. An encoded bitstream generally includes a set of values ​​for syntax elements that represent coding decisions (e.g., coding modes).

[0029] This disclosure may generally refer to "signaling" some information, such as a syntax element. The term "signaling" may generally refer to communication of values ​​of syntax elements and / or other data used to decode encoded data. That is, the G-PCC encoder 200 may signal values ​​of syntax elements in a bitstream. In general, signaling refers to generating values ​​in a bitstream. As noted above, the source device 102 may transport the bitstream to the destination device 116 in substantially real-time or non-real-time, which may occur, such as when storing syntax elements in the storage device 112 for later retrieval by the destination device 116.

[0030] ISO / IEC MPEG (JTC1 / SC29 / WG11) is investigating the potential need for, and aims to develop, a standard for, point cloud coding techniques with compression capabilities significantly beyond those of current methods. The group is working together on this research in a collaborative effort known as the 3-Dimensional Graphics Team (3DG) to evaluate compression design proposals proposed by experts in the field.

[0031] Point cloud compression activities are categorized into two different approaches. The first approach is "Video Point Cloud Compression" (V-PCC), which segments a 3D object and projects the segments into multiple 2D planes (represented as "patches" in a 2D frame), which are further coded by a legacy 2D video codec, such as the High Efficiency Video Coding (HEVC) (ITU-T H.265) codec. The second approach is "Geometry-Based Point Cloud Compression" (G-PCC), which directly compresses the 3D geometry, i.e., the location of a set of points in 3D space and the associated attribute values ​​(for each point related to the 3D geometry). G-PCC addresses the compression of point clouds in both category 1 (static point clouds) and category 3 (dynamically acquired point clouds). A draft of the G-PCC standard is available in G-PCC DIS, ISO / IEC JTC1 / SC29 / WG11 w19522, MPEG-131, Teleconference, July 2020, and a codec description is available in G-PCC Codec Description, ISO / IEC JTC1 / SC29 / WG11 w19525, MPEG-131, Teleconference, July 2020.

[0032] The units shown are logical and do not necessarily correspond one-to-one with the code implemented in the reference implementation of the G-PCC codec, i.e., the TMC13 test model software studied by ISO / IEC MPEG (JTC1 / SC29 / WG11).Similarly, the units shown do not necessarily correspond one-to-one with the hardware units in the hardware implementation of the G-PCC codec.

[0033] In both the G-PCC encoder 200 and the G-PCC decoder 300, the location of the point cloud is coded first. The attribute coding depends on the decoded geometry. For category 3 data, the compressed geometry is typically represented as an octree from the root to the leaf level of the individual voxels. For category 1 data, the compressed geometry is typically represented by a pruned octree (i.e. an octree from the root to the leaf level of a block larger than a voxel) plus a model that approximates the surface in each leaf of the pruned octree. In this way, both category 1 and category 3 data share the octree coding mechanism, and the category 1 data may further approximate the voxels in each leaf with a surface model. The surface model used is a triangulation with 1 to 10 triangles per block, resulting in a triangle soup. Thus, category 1 geometry codecs are known as Trisoup geometry codecs, and category 3 geometry codecs are known as Octree geometry codecs.

[0034] At each node of the octree, occupancy is signaled (when not inferred) for one or more of its child nodes (up to a maximum of eight nodes). Multiple neighborhoods are specified, including (a) nodes that share a face with the current octree node, (b) nodes that share a face, edge, or vertex with the current octree node, etc. Within each neighborhood, the occupancy of a node and / or its children may be used to predict the occupancy of the current node or its children. For sparsely distributed points in several nodes of the octree, the codec also supports a direct coding mode in which the 3D location of the points is directly coded. A flag may be signaled to indicate that the direct mode is signaled. At the lowest level, the number of points associated with an octree node / leaf node may also be coded.

[0035] When geometry is coded, attributes corresponding to the geometry points are coded. When there are multiple attribute points corresponding to one reconstructed / decoded geometry point, an attribute value representing the reconstructed point may be derived.

[0036] In G-PCC, there are three attribute coding methods: Region Adaptive Hierarchical Transform (RAHT) coding, Interpolation-based Hierarchical Nearest Neighbor Prediction (Predicting Transform), and Interpolation-based Hierarchical Nearest Neighbor Prediction with Update / Lifting Steps (Lifting Transform). RAHT and Lifting are typically used for Category 1 data, and Predicting is typically used for Category 3 data. However, either method may be used for any data, and as in the geometry codec in G-PCC, the attribute coding method used to code the point cloud is specified in the bitstream.

[0037] The coding of attributes may be done at a level of detail (LOD), where each level of detail may be used to obtain a more refined representation of the point cloud attributes. Each level of detail may be specified based on a distance measure from neighboring nodes or based on a sampling distance.

[0038] The G-PCC encoder 200 may quantize the residual obtained as the output of the coding method for the attribute. The G-PCC encoder 200 may entropy code the quantized residual using context-adaptive arithmetic coding.

[0039] According to the techniques of the present disclosure, the G-PCC encoder 200 and the G-PCC decoder 300 may be configured to separately encode / decode the points of the point cloud based on the classification of the points. Specifically, the G-PCC encoder 200 and the G-PCC decoder 300 may be configured to classify the points, for example, into ground (or road) points and object points. In some examples, a LIDAR system mounted on a car may project a laser into the surrounding environment to build a point cloud. The present disclosure recognizes that the ground or road on which the car is traveling is likely to remain relatively flat and stable from frame to frame (i.e., between each instance of point cloud construction). Therefore, the points collected at the ground or road locations should be approximately identical between each frame.

[0040] For other parts of the point cloud, the identified points may correspond to non-road / ground objects. Thus, the relative location of each of the points corresponding to non-road / ground objects may change from frame to frame in substantially the same manner due to the speed of the vehicle. Therefore, it may be efficient to encode and decode the points corresponding to the objects using a global motion vector, and encode and decode the points corresponding to the road or ground using a different mechanism, for example, a different global motion vector (such as a zero-valued global motion vector), a local motion vector, or intra prediction.

[0041] The G-PCC encoder 200 may determine thresholds for classifying points as either ground / road points (hereafter generally referred to as "ground" points) or object points. For example, the G-PCC encoder 200 may determine a top threshold and a bottom threshold that generally represent the top and bottom of the ground or road. Thus, if a point is between these two thresholds, the point may be classified as a ground point, and other points (points above the top threshold or below the bottom threshold) may be classified as object points. The G-PCC encoder 200 may encode data representing the top and bottom thresholds in a data structure, such as a sequence parameter set (SPS), a geometry parameter set (GPS), or a geometry data unit header (GDH). Thus, the G-PCC encoder 200 and the G-PCC decoder 300 may use a second different global motion vector, local motion vector, intra prediction, or other different prediction technique to encode or decode the occupancy of nodes above the top threshold or below the bottom threshold using global motion vectors and nodes between the top and bottom thresholds.

[0042] In this manner, the techniques of this disclosure may result in more efficient coding of object points. Instead of coding points in point clouds using their respective local motion vectors, all of the object points between each group may be predicted using a single global motion vector. Thus, the signaling overhead associated with signaling motion information for object points may be significantly reduced. Moreover, since ground points may be largely assumed to remain constant between frames, the coding techniques for ground points may consume relatively few bits.

[0043] FIG. 2 is a block diagram illustrating exemplary components of the G-PCC encoder 200 of FIG. 1 that may be configured to perform the techniques of this disclosure. In the example of FIG. 2, the G-PCC encoder 200 includes a memory 228, a coordinate transformation unit 202, a color transformation unit 204, a voxelization unit 206, an attribute transfer unit 208, an octree analysis unit 210, a surface approximation analysis unit 212, an arithmetic coding unit 214, a geometry reconstruction unit 216, a RAHT unit 218, an LOD generation unit 220, a lifting unit 222, a coefficient quantization unit 224, and an arithmetic coding unit 226. In FIG. 2, the gray shaded units are options that are typically used for category 1 data. In FIG. 2, the diagonal lined units are options that are typically used for category 3 data. All other units are common between categories 1 and 3.

[0044] As shown in the example of Figure 2, the G-PCC encoder 200 may receive a set of locations and a set of attributes. The locations may include coordinates of points in the point cloud. The attributes may include information about the points in the point cloud, such as a color associated with the points in the point cloud.

[0045] The coordinate transformation unit 202 may apply a transformation to the coordinates of the points to convert the coordinates from an initial domain to a transformation domain. In this disclosure, the transformed coordinates may be referred to as transformed coordinates. The color transformation unit 204 may apply a transformation to convert color information of the attributes to a different domain. For example, the color transformation unit 204 may convert the color information from an RGB color space to a YCbCr color space.

[0046] Further, in the example of FIG. 2, the voxelization unit 206 may voxelize the transformed coordinates. The voxelization of the transformed coordinates may include quantization and removing some points of the point cloud. In other words, multiple points of the point cloud may be contained within a single "voxel," which may then be treated as one point in some respects. The octree analysis unit 210 may also store data representing occupied voxels (i.e., voxels occupied by points of the point cloud) in the memory 228 (e.g., in a history buffer of the memory 228).

[0047] Additionally, the arithmetic coding unit 214 may entropy code data representing the occupancy of the octree. In some examples, the arithmetic coding unit 214 may entropy code the occupancy data based only on data of the current point cloud (sometimes referred to as "intra prediction" of the current point cloud). In other examples, the arithmetic coding unit 214 may entropy code the occupancy data with reference to a previous octree for a previous point cloud, e.g., buffered in memory 228 (sometimes referred to as "inter prediction" of the current point cloud relative to a reference cloud). As discussed in more detail below, e.g., with respect to FIG. 3, the arithmetic coding unit 214 may perform inter prediction using local or global motion vectors.

[0048] Specifically, according to the techniques of this disclosure, the arithmetic coding unit 214 may entropy decode data representing thresholds (e.g., top and bottom thresholds) for defining ground points (or road points) and object points. The top and bottom thresholds may correspond to a series of frames (point cloud). The arithmetic coding unit 214 may also entropy decode data representing a global motion vector for a current point cloud of the series of frames. The arithmetic coding unit 214 may form a predicted cloud using a global motion vector from a previous point cloud buffered in memory 228, and use the occupancy of nodes in the predicted cloud to determine a context for entropy decoding occupancy data of nodes that are either above the top threshold or below the bottom threshold of the current cloud. The arithmetic coding unit 214 may use different prediction techniques for ground / road points, such as different global motion vectors, local motion vectors, intra prediction, or another alternative entropy decoding / prediction technique.

[0049] 2, a surface approximation analysis unit 212 may analyze the points to possibly determine a surface representation of the set of points. An arithmetic coding unit 214 may entropy code syntax elements representing the octree and / or surface information determined by the surface approximation analysis unit 212. The G-PCC encoder 200 may output these syntax elements in a geometry bitstream.

[0050] The geometry reconstruction unit 216 may reconstruct transformation coordinates of points in the point cloud based on the octree, the data indicative of the surface determined by the surface approximation analysis unit 212, and / or other information. The number of transformation coordinates reconstructed by the geometry reconstruction unit 216 may differ from the original number of points in the point cloud due to voxelization and surface approximation. In this disclosure, the resulting points may be referred to as reconstructed points. The attribute transfer unit 208 may transfer attributes of the original points of the point cloud to the reconstructed points of the point cloud.

[0051] Further, the RAHT unit 218 may apply RAHT coding to the attributes of the reconstructed points. Alternatively or additionally, the LOD generation unit 220 and the lifting unit 222 may apply LOD processing and lifting, respectively, to the attributes of the reconstructed points. The RAHT unit 218 and the lifting unit 222 may generate coefficients based on the attributes. The coefficient quantization unit 224 may quantize the coefficients generated by the RAHT unit 218 or the lifting unit 222. The arithmetic coding unit 226 may apply arithmetic coding to syntax elements representing the quantized coefficients. The G-PCC encoder 200 may output these syntax elements in an attribute bitstream.

[0052] FIG. 3 is a conceptual diagram illustrating an example of inter-prediction encoding in G-PCC. In some examples, the G-PCC encoder 200 may decode / replicate a point cloud to form a reference group 130. In other examples, the G-PCC encoder 200 may simply store an unencoded past version of a previous point cloud. The reference group 130 may be stored in a decoded frame buffer or history buffer (i.e., memory) of the G-PCC encoder 200. The G-PCC encoder 200 may further obtain a current group 140 that is to be encoded, at least in part, using inter prediction. For example, the G-PCC encoder 200 may use techniques of this disclosure to determine a set of points of the current group 140 that are to be predicted using global motion rather than local motion or intra prediction.

[0053] The G-PCC encoder 200 may compare the positions of the points of the current set 140 to be inter predicted with the points of the reference set 130 and calculate a global motion vector 132. The global motion vector 132 may represent a global motion vector that most accurately predicts the relative position of the points of the current set to be inter predicted using global motion with respect to the reference set 130. The G-PCC encoder 200 may then form a predicted set 134 by applying the global motion vector 132 to the reference set 130. That is, the G-PCC encoder 200 may construct the predicted set 134 by applying the global motion vector 132 to each point of the reference set 130 at their respective positions and setting the occupancy of the nodes to include the points in the predicted set 134 at the corresponding positions offset by the global motion vector 132.

[0054] The G-PCC encoder 200 (and in particular the arithmetic coding unit 214) may then encode the points of the nodes of the current set 140 using the corresponding points in the nodes of the predicted set 134 to determine a context for context-based entropy coding, e.g., context-adaptive binary arithmetic coding (CABAC). For example, the arithmetic coding unit 214 may encode the occupancy of the current node 142 of the current set 140 using the occupancy of the reference node 136 (corresponding to the position of the current node 142 as indicated by the vector 144) to determine a context for encoding the value of the occupancy of the current node 142.

[0055] For example, if the reference node 136 is occupied (i.e., includes a point), the arithmetic coding unit 214 may determine a first context for encoding a value representative of the occupancy of the current node 142. The first context may indicate a most probable symbol for the value representative of the occupancy of the current node 142 as having a high probability of a value representative of the current node 142 being occupied (e.g., "1"). On the other hand, if the reference node 136 is not occupied (i.e., does not include any point), the arithmetic coding unit 214 may determine a second context for encoding a value representative of the occupancy of the current node 142. The second context may indicate a most probable symbol for the value representative of the occupancy of the current node 142 as having a high probability of a value representative of the current node 142 being unoccupied (e.g., "0"). The arithmetic coding unit 142 may then determine whether the current node 142 is actually occupied, determine a value representing whether the current node 142 is actually occupied, and then entropy code the value using the determined context (e.g., the first context or the second context). The arithmetic coding unit 214 may add the entropy coded value to the bitstream 146 and proceed to the next node in the current group 140 (or the next group).

[0056] FIG. 4 is a block diagram illustrating exemplary components of the G-PCC decoder 300 of FIG. 1 that may be configured to perform the techniques of this disclosure. In the example of FIG. 4, the G-PCC decoder 300 includes a geometry arithmetic decoding unit 302, a memory 324, an attribute arithmetic decoding unit 304, an octree synthesis unit 306, an inverse quantization unit 308, a surface approximation synthesis unit 310, a geometry reconstruction unit 312, a RAHT unit 314, a LoD generation unit 316, an inverse lifting unit 318, an inverse coordinate transformation unit 320, and an inverse color transformation unit 322. In FIG. 4, the gray shaded units are options that are typically used for category 1 data. In FIG. 4, the diagonal lined units are options that are typically used for category 3 data. All other units are common between categories 1 and 3.

[0057] The G-PCC decoder 300 may obtain a geometry bitstream and an attribute bitstream. A geometry arithmetic decoding unit 302 of the decoder 300 may apply arithmetic decoding (e.g., context-adaptive binary arithmetic coding (CABAC) or other types of arithmetic decoding) to syntax elements in the geometry bitstream. Similarly, an attribute arithmetic decoding unit 304 may apply arithmetic decoding to syntax elements in the attribute bitstream.

[0058] The geometry arithmetic decoding unit 302 may entropy decode data representing the occupancy of the octree for the current point cloud. In some examples, the geometry arithmetic decoding unit 302 may entropy decode the occupancy data based only on the data of the current point cloud (sometimes referred to as "intra prediction" of the current point cloud). In other examples, the geometry arithmetic decoding unit 302 may entropy decode the occupancy data with reference to a previous octree for a previous point cloud, e.g., buffered in memory 324 (sometimes referred to as "inter prediction" of the current point cloud relative to a reference cloud). As discussed in more detail below, e.g., with respect to FIG. 5, the geometry arithmetic decoding unit 302 may perform inter prediction using local or global motion vectors.

[0059] Specifically, according to the techniques of this disclosure, the geometry arithmetic decoding unit 302 may entropy decode data representing thresholds (e.g., top and bottom thresholds) for defining ground points (or road points) and object points. The top and bottom thresholds may correspond to a series of frames (point cloud). The geometry arithmetic decoding unit 302 may also entropy decode data representing a global motion vector for a current point cloud of the series of frames. The geometry arithmetic decoding unit 302 may form a predicted cloud using a global motion vector from a previous point cloud buffered in memory 324, and use the occupancy of nodes in the predicted cloud to determine a context for entropy decoding occupancy data of nodes that are either above the top threshold or below the bottom threshold of the current cloud. The geometry arithmetic decoding unit 302 may use a different prediction technique for ground / road points, such as a different global motion vector, local motion vector, intra prediction, or another alternative entropy decoding / prediction technique.

[0060] The octree synthesis unit 306 may synthesize an octree based on data for syntax elements parsed from the geometry bitstream and entropy decoded by the geometry arithmetic decoding unit 302. In cases where surface approximations are used in the geometry bitstream, the surface approximation synthesis unit 310 may determine a surface model based on the syntax elements parsed from the geometry bitstream and based on the octree.

[0061] Additionally, the geometry reconstruction unit 312 may perform reconstruction to determine coordinates of points in the point cloud. The inverse coordinate transformation unit 320 may apply an inverse transformation to the reconstructed coordinates to transform the reconstructed coordinates (locations) of points in the point cloud from the transformed domain back to the initial domain.

[0062] 4, the inverse quantization unit 308 may inverse quantize the attribute value. The attribute value may be based on syntax elements obtained from the attribute bitstream (e.g., including syntax elements decoded by the attribute arithmetic decoding unit 304).

[0063] Depending on how the attribute values ​​are encoded, the RAHT unit 314 may perform RAHT coding to determine color values ​​for the points of the point cloud based on the dequantized attribute values. Alternatively, the LOD generation unit 316 and the inverse lifting unit 318 may use level-of-detail based techniques to determine color values ​​for the points of the point cloud.

[0064] 4, the inverse color transformation unit 322 may apply an inverse color transformation to the color values. The inverse color transformation may be the inverse of the color transformation applied by the color transformation unit 204 of the encoder 200. For example, the color transformation unit 204 may transform the color information from the RGB color space to the YCbCr color space. Thus, the inverse color transformation unit 322 may transform the color information from the YCbCr color space to the RGB color space.

[0065] The various units shown in FIG. 2 and FIG. 4 are shown to aid in understanding the operations performed by the encoder 200 and the decoder 300. The units may be implemented as fixed function circuits, programmable circuits, or a combination thereof. Fixed function circuits refer to circuits that provide a specific function, and the operations that may be performed are predefined. Programmable circuits refer to circuits that may be programmed to perform various tasks, and provide flexibility in the operations that may be performed. For example, a programmable circuit may execute software or firmware that causes the programmable circuit to operate in a manner defined by the software or firmware instructions. Although a fixed function circuit may execute software instructions (e.g., to receive parameters or output parameters), the type of operations that the fixed function circuit performs is largely invariant. In some examples, one or more of the units may be separate circuit blocks (fixed function or programmable), and in some examples, one or more of the units may be an integrated circuit.

[0066] 5 is a conceptual diagram illustrating an example of inter-prediction decoding in G-PCC. According to the techniques of this disclosure, the G-PCC decoder 300 may decode a set of points of the current group 160 using the global motion vector inter prediction technique of FIG. 5 and decode a second set of points of the current group 160 using local motion vector inter prediction or intra prediction. The G-PCC decoder 300 may receive and decode data of the bitstream 166 indicating whether a set of points for one or more nodes should be decoded using global motion vector inter prediction.

[0067] The G-PCC decoder 300 may first decode one or more previous point clouds and store the previously decoded point clouds in a decoding frame buffer or history buffer (i.e., the memory of the G-PCC decoder 300). The G-PCC decoder 300 may also decode motion information, including data for the global motion vector 152 and identifying the reference group 150 in the previously decoded point clouds.

[0068] The G-PCC decoder 300 may apply the global motion vector 152 to the reference set 150 to generate a predicted set 154. That is, the G-PCC decoder 300 may construct the predicted set 154 by applying the global motion vector 152 to each point of the reference set 150 at their respective positions and setting the occupancy of a node (e.g., the reference node 156) of the predicted set 154 to include a point at a corresponding position that is offset by the global motion vector 152.

[0069] The geometry arithmetic decoding unit 302 may then use the occupancy of a node (e.g., the reference node 156) of the predicted group 154 ​​to determine a context for decoding a value representative of the occupancy of the current node 162 of the current group 160. The current node 162 corresponds to the reference node 156 as indicated by the vector 164. For example, if the reference node 156 is occupied (i.e., contains a point), the geometry arithmetic decoding unit 302 may determine a first context for encoding a value representative of the occupancy of the current node 162. The first context may indicate a highest probability symbol for a value representative of the occupancy of the current node 162 as having a high probability of a value (e.g., "1") representing that the current node 162 is occupied. On the other hand, if the reference node 156 is not occupied (i.e., does not contain any point), the geometry arithmetic decoding unit 302 may determine a second context for encoding a value representative of the occupancy of the current node 162. The second context may indicate the highest probability symbol for a value representing the occupancy of the current node 162 as having a high probability of a value (e.g., “0”) representing that the current node 162 is unoccupied. The geometry arithmetic decoding unit 302 may then decode the value of the bitstream 166 representing the occupancy of the current node 162 using the determined context.

[0070] 6 is a conceptual diagram illustrating an example prediction tree that may be used when performing the techniques of this disclosure. As an alternative to octree geometry coding, predictive geometry coding was introduced in "Exploratory model for inter-prediction in G-PCC", ISO / IEC JTC1 / SC29 WG11, Document N18096, Macau, China, October 2018. In predictive geometry coding, nodes are arranged in a tree structure (which defines a prediction structure), and various prediction strategies are used to predict the coordinates of each node in the tree with respect to the predictors of the tree.

[0071] FIG. 6 illustrates an exemplary prediction tree 350, which is a directed graph with arrows pointing in the prediction direction. Prediction tree 350 includes different types of nodes according to the number of children (e.g., 0 to 3). In the example of FIG. 6, node 352 is an example of a branch vertex with three children, node 354 is an example of a branch node with two children, node 356 is an example of a branch node with one child, node 358 represents an example of a leaf vertex, and node 360 ​​represents an example of a root vertex. As a root vertex, node 360 ​​has no predictors. Each node in prediction tree 350 has at most one parent node.

[0072] For the current node, four prediction strategies may be defined based on its parent (p0), grandparent (p1), and great-grandparent (p2): 1) no prediction / zero prediction (0), 2) delta prediction (p0), 3) linear prediction (2*p0-p1), parallelogram prediction (2*p0+p1-p2).

[0073] The G-PCC encoder 200 may use any algorithm to generate a prediction tree. The G-PCC encoder 200 may determine the algorithm to be used according to the application / use case, and several strategies may be used. Some strategies are described in N18096.

[0074] For each node, the G-PCC encoder 200 may start encoding the residual coordinate values ​​in the bitstream from the root node in a depth-first manner. Predictive geometry coding may be particularly useful for category 3 (e.g., LIDAR-acquired) point cloud data, e.g., for low-latency applications.

[0075] FIG. 7 is a conceptual diagram illustrating an exemplary rotational LIDAR acquisition model. FIG. 7 shows a LIDAR 380 including several sensors emitting and receiving respective lasers 382. G-PCC includes an angular mode for predictive geometry coding. In the angular mode, the characteristics of the LIDAR sensor may be used to code the prediction tree more efficiently. In the angular mode, the location coordinates are converted into radius (r) 384, azimuth (Φ) 386, and laser index (i) 388 values. The G-PCC encoder 200 and the G-PCC decoder 300 may perform prediction in this domain. That is, the G-PCC encoder 200 and the G-PCC decoder 300 may code the residual values ​​in the r, Φ, i domain.

[0076] Due to rounding errors, coding in the r, Φ, i domain is not lossless. Therefore, the G-PCC encoder 200 and the G-PCC decoder 300 may code a second set of residuals corresponding to the Cartesian coordinates. A description of the encoding and combining strategy used for the angular mode of predictive geometry coding is reproduced below from N18096.

[0077] The angular mode of predictive geometry coding focuses on point clouds acquired using a rotating Lidar model. In the example of FIG. 7, the LIDAR 380 has N lasers 382 (e.g., N may be equal to 16, 32, 64, or some other value) that rotate around the Z axis according to an azimuth angle Φ. Each of the lasers 382 is focused at a different elevation angle θ(i). i=1...N and height σ(i) i=1…N Let laser i strike a point M with orthogonal integer coordinates (x, y, z) defined according to some coordinate system.

[0078] According to N18096, the location of a point M may be modeled using three parameters (r, φi), which are

number

[0079] More precisely, the G-PCC encoder 200 and the G-PCC decoder 300

number

number

number

number

[0080] To avoid reconstruction discrepancies due to the use of floating-point arithmetic, the G-PCC encoder 200 and the G-PCC decoder 300 calculate σ(i) as follows: i=1…N and tan(θ(i)) i=1…N The values ​​of may be pre-calculated and quantized.

number

number

[0081] The G-PCC encoder 200 and the G-PCC decoder 300 may obtain the reconstructed Cartesian coordinates as follows:

number

[0082] For various reasons, such as quantization, approximation, model inaccuracies, and / or model parameter inaccuracies,

number

[0083] (r x ,r y ,r z ) be the reconstruction residual, defined as

number

[0084] The G-PCC encoder 200 may proceed as follows. Model parameters

number

number

number

number

[0085] The G-PCC decoder 300 may proceed as follows. Model parameters

number

number

number

number

[0086] The G-PCC encoder 200 reconstructs the residual (r x ,r y ,r z ), or omitting points, lossy compression may be achieved. The quantized reconstructed residual may be calculated as:

number

number

[0087] To further improve the RD (rate-distortion) performance results, lattice quantization may be used. To achieve domain-adaptive quality and for rate control purposes, the quantization parameters may be varied at the sequence / frame / slice / block level.

[0088] FIG. 8 is a flow chart showing an example motion estimation process for the G-PCC InterEM software. There are two types of motion involved in the G-PCC InterEM software: global motion matrix and local nodal motion vectors. Global motion parameters are defined as a rotation matrix and a translation vector, which are applied to all points in the predicted (reference) frame. Local nodal motion vectors of a node of an octree are motion vectors that may only be applied to points within the node in the predicted (reference) frame. Details of the motion estimation algorithm in InterEM are described below. FIG. 8 shows a flow chart of the motion estimation algorithm.

[0089] Given an input predicted (reference) frame and a current frame, the G-PCC encoder 200 may first estimate global motion at a global scale (400). The G-PCC encoder 200 may then apply the estimated global motion to the predicted (reference) frame (402). After applying global motion to the predicted (reference) frame, the G-PCC encoder 200 may estimate local motion at a finer scale, for example at the node level of the octree (404). Finally, the G-PCC encoder 200 may perform motion compensation to encode the estimated local node motion vectors and points (406).

[0090] FIG. 9 is a flow chart showing an exemplary process for estimating global motion. In the InterEM software, a global motion matrix is ​​defined to match feature points between a predicted frame (reference) and a current frame. FIG. 9 shows a pipeline for estimating global motion. The global motion estimation algorithm may be divided into three steps: feature point discovery (410), feature point pair sampling (412), and motion estimation using a least mean squares (LMS) algorithm (414).

[0091] The algorithm defines feature points as points with large location changes between the predicted frame and the current frame. For each point in the current frame, the G-PCC encoder 200 finds the closest point in the predicted frame and pairs points between the current frame and the predicted frame. If the distance between the paired points is greater than a threshold, the G-PCC encoder 200 considers the paired points as feature points.

[0092] After finding the feature points, the G-PCC encoder 200 performs sampling on the feature points to reduce the problem size (e.g., by selecting a subset of feature points to reduce the complexity of motion estimation).The G-PCC encoder 200 then applies the LMS algorithm to derive motion parameters by attempting to reduce the error between each feature point in the predicted frame and the current frame.

[0093] 10 is a flowchart illustrating an example process for estimating local node motion vectors. The G-PCC encoder 200 may recursively estimate motion vectors for nodes of a prediction tree. The G-PCC encoder 200 may evaluate a cost function to select an optimal motion vector based on a rate-distortion (RD) cost.

[0094] In the example of FIG. 10, the G-PCC encoder 200 receives a current node (420). If the current node is not split into eight children, the G-PCC encoder 200 determines the motion vector with the lowest cost between the current node and the prediction node (422). On the other hand, if the current node is split into eight children, the G-PCC encoder 200 splits the current node into eight children (424), finds the motion of each child node (426), and adds all the returned estimated costs (428). That is, the G-PCC encoder 200 obtains the total cost under the split condition by applying a motion estimation algorithm and adding the estimated cost value of each child node. The G-PCC encoder 200 may determine whether to split the node by comparing the cost with and without splitting. If split, the G-PCC encoder 200 may assign each partial node a respective motion vector (or further split the node into respective child nodes). If not, the G-PCC encoder 200 may assign the node's motion vector to the node. The G-PCC encoder 200 may then compare the costs to determine whether to split the current node or not to split the current node (430).

[0095] Two parameters that can affect the performance of motion vector estimation are block size (BlockSize) and minimum prediction unit size (MinPUSize). BlockSize defines the upper limit and MinPUSize defines the lower limit of the node size to which motion vector estimation should be applied.

[0096] US Provisional Application No. 63 / 090657, filed October 12, 2020, which was converted to US Patent Application No. 17 / 495,428, filed October 2021, describes an improved global motion estimation technique based on an Iterative Closest Point method. In this method, an initial translation vector is first estimated by minimizing the mean square error between the current frame and the reference frame. When estimating the initial translation vector, a label of whether the point is a ground or not may be taken into account. For example, if the point is a ground point, the point is excluded from the estimation. The initial translation vector combined with the identity matrix may then be fed into an Iterative Closest Point method or a similar method to estimate a rotation matrix and a translation vector. Also, in this case, whether the point is a ground or not may be taken into account, for example, by excluding the point from the estimation. Alternatively, a rotation matrix may be first estimated based on a label of whether the point is a ground or not. This label may be derived by the G-PCC encoder 200 and signaled to the G-PCC decoder 300, or the G-PCC encoder 200 and the G-PCC decoder 300 may derive the label. This label may be derived based on a ground estimation algorithm, which may be based on the height of the point, the density of the point cloud near the point, the relative distance of the point from the LIDAR origin / fixed point, etc.

[0097] In real-world applications such as automobiles, ground areas and objects in a point cloud usually have different motions. For example, ground points may have zero or little motion, while objects may have larger motion. In traditional methods for estimating global motion in InterEM software, both ground points and object points may be used to derive global motion. After doing so, the estimation output may not be accurate.

[0098] U.S. Provisional Patent Application No. 63 / 090657, filed October 12, 2020, introduced several labeling methods for classifying objects and ground. For example, in these methods, the G-PCC encoder 200 may derive a label and signal the label to the G-PCC decoder 300, or both the G-PCC encoder 200 and the G-PCC decoder 300 may derive a label. The label may be derived based on a ground estimation algorithm, which may be based on the height of the point, the density of the point cloud near the point, and the relative distance of the point from the LIDAR origin / fixed point.

[0099] This disclosure describes techniques for labeling ground and objects to improve the performance of global motion estimation. Specifically, the G-PCC encoder 200 and the G-PCC decoder 300 may be configured to classify ground / road and object data in a point cloud, which may improve the performance of global motion estimation.

[0100] 11 is a graph illustrating an example classification of a cloud into ground (road) and objects using two thresholds of the z-values ​​of the points according to the techniques of this disclosure. The G-PCC encoder 200 and the G-PCC decoder 300 may be configured to classify the ground and road points using the heights (or z-values) of the points in the cloud. In one example, the G-PCC encoder 200 and the G-PCC decoder 300 may be configured with a two-threshold definition, e.g., z_top 452 and z_bottom 454 as shown in FIG. 11.

[0101] If the height (z value) of the point is lower than z_bottom 454 or higher than z_top 452, the G-PCC encoder 200 and the G-PCC decoder 300 may classify the point as an object. Otherwise, if the height (z value) of the point is between z_bottom 454 and z_top 452, the G-PCC encoder 200 and the G-PCC decoder 300 may classify the point as ground (road).

[0102] In some examples, the G-PCC encoder 200 and the G-PCC decoder 300 may specify a ground point using a set of value ranges and classify the ground point as including any point that satisfies at least one of the value ranges. For example, for a coordinate of (x, y, z), the G-PCC encoder 200 and the G-PCC decoder 300 may classify the ground point as including any point that satisfies at least one of the value ranges. i , x_max i ), (y_min i , y_max i ), (z_min i , z_max i )}. The G-PCC encoder 200 and the G-PCC decoder 300 may be configured with N such ranges, where i is in [1,N]. The G-PCC encoder 200 and the G-PCC decoder 300 may be configured with a range of i-th values, such that ((x_min i ≦x≦x_max i ) & (y_min i ≦y≦y_max i ) & (z_min i ≦z≦z_max i )), then we may classify the point at (x,y,z) as a ground point.

[0103] The G-PCC encoder 200 and the G-PCC decoder 300 may be configured with a minimum value up to negative infinity and a maximum value up to infinity. In the above example for the z-values ​​of points between Z_bottom 454 and z_top 452 that are classified as ground, and other points that are classified as objects, x_min i and y_min i may be set to negative infinity, and x_max i and y_max i may be set to infinity, but z_min i z_bottom may be set to 454, z_max i may be set to z_top 452.

[0104] When the G-PCC encoder 200 is configured to quantize the point cloud before encoding by a scaling factor, the G-PCC encoder 200 and the G-PCC decoder 300 may also quantize the thresholds using the same quantization factor.

[0105] Additionally or alternatively, the G-PCC encoder 200 and the G-PCC decoder 300 may be configured to use the output of the point classification (e.g., into road and ground points) in global motion estimation and prediction. As discussed above, the G-PCC encoder 200 and the G-PCC decoder 300 may estimate global motion according to the methods described in the InterEM Software techniques, U.S. Provisional Application No. 63 / 090,657.

[0106] Alternatively, in some examples, the G-PCC encoder 200 and the G-PCC decoder 300 may derive two global motion sets. The G-PCC encoder 200 and the G-PCC decoder 300 may use a first set of global motion estimates to predict ground / road points and a second set to predict object points. To derive the global motion set for ground / road, only points labeled "ground / road" may be used. To derive the global motion set for objects, only points labeled "object" may be used.

[0107] As yet another example, the G-PCC encoder 200 and the G-PCC decoder 300 may derive only one global motion set to predict object points. In this example, the G-PCC encoder 200 and the G-PCC decoder 300 may predict ground / road points using zero motion (translation and rotation set equal to 0).

[0108] In some examples, which may be used in addition to the various techniques discussed above, the G-PCC encoder 200 and the G-PCC decoder 300 may define thresholds for different levels of sharing. For example, the G-PCC encoder 200 and the G-PCC decoder 300 may determine thresholds independently for different frames. The G-PCC encoder 200 may determine a threshold for each frame and encode data representing the threshold in the bitstream, so that the G-PCC decoder 300 can determine the threshold from the encoded data of the bitstream.

[0109] In some examples, the G-PCC encoder 200 and the G-PCC decoder 300 may define the threshold at a group of pictures (GOP) level. In this example, all frames in a GOP may share the same threshold. The G-PCC encoder 200 and the G-PCC decoder 300 may determine the shared threshold at the beginning of the GOP and code the data together with encoding information for the first frame of the GOP.

[0110] In some examples, the G-PCC encoder 200 and the G-PCC decoder 300 may define the threshold at a sequence level, i.e., all frames in a sequence may share the same threshold. In this example, the threshold may be presented in encoder configuration data (e.g., an encoder configuration file) for the G-PCC encoder 200, and the G-PCC encoder 200 may encode data representing the threshold in the bitstream so that the G-PCC decoder 300 can determine the threshold from the encoded data.

[0111] The G-PCC encoder 200 may use various techniques to derive the thresholds that apply to a set of two or more frames (eg, a GOP, a sequence, etc.). In the simplest case, the G-PCC encoder 200 may select the threshold value of the first frame in the set as the threshold value for the frames in the set, which may be the first frame in the output order or decoding order of the point cloud. In some examples, the G-PCC encoder 200 may derive the threshold to be used according to a weighted average of thresholds derived / applicable for two or more frames in the set. For example, if there are 10 frames in the set and t1 i , t2 i If denotes the threshold value derived for the i-th frame, then for n equal to 1 and 2, the final threshold value may be derived as follows:

number

[0112] In some examples, the G-PCC encoder 200 may derive the thresholds to encode data in a bitstream so that the G-PCC decoder 300 can determine the thresholds from the encoded data. In some examples, both the G-PCC encoder 200 and the G-PCC decoder 300 may derive the thresholds according to the same technique.

[0113] In some examples, when the G-PCC encoder 200 signals the threshold, the G-PCC encoder 200 may signal the threshold in a sequence parameter set (SPS), a geometry parameter set (GPS), or a geometry data unit header (GDH). In one example where the G-PCC encoder 200 signals the threshold at the GPS / GDH level, the G-PCC encoder 200 may be configured to conditionally signal the threshold, for example, only when the angle mode is enabled. Thus, when the angle mode is enabled, the G-PCC encoder 200 and the G-PCC decoder 300 may be configured to code data for the threshold in the GPS / GDH, while when the angle mode is disabled, the G-PCC encoder 200 and the G-PCC decoder 300 may be configured to avoid coding data from the threshold. Alternatively, the G-PCC encoder 200 and the G-PCC decoder 300 may be configured to unconditionally code data for the threshold.

[0114] The G-PCC encoder 200 and the G-PCC decoder 300 may code the data for the threshold as a value of se(v) or ue(v). Se(v) coding may represent a left bit-first signed integer zeroth exponential Golomb coded syntax element, while ue(v) coding may represent a left bit-first unsigned integer zeroth exponential Golomb coded syntax element. In one example, the GPS may be modified in Table 1 as follows, where [addition:"text to be added"] represents an addition to the existing G-PCC standard, and the G-PCC encoder 200 and the G-PCC decoder 300 may be configured accordingly. [Table 1]

[0115] Alternatively, if the threshold is large enough, a fixed length coding can also be performed, which involves indicating the number of bits to be coded for the fixed length coding, followed by the actual fixed length coding of the threshold using s(v) coding, e.g., according to Table 2 below, where s(v) coding represents a signed fixed length coding of the value. [Table 2]

[0116] As shown in FIG. 11, there may be a maximum of two thresholds (z_bottom, z_top) in the scenarios of top region, road region, and bottom region. In a typical scenario, the origin of the frame may be the center of the LIDAR system, which indicates that both z_top and z_bottom are likely to be negative since the center of the LIDAR system / origin of the frame is likely to be well above the road. Secondly, the thresholds may always be ordered in descending order, i.e. z_top>z_bottom. In such a case, the G-PCC encoder 200 and the G-PCC decoder 300 may be configured to code the first threshold as is, and for the second threshold, the G-PCC encoder 200 and the G-PCC decoder 300 may code the difference between the second threshold and the first threshold. Moreover, since this delta is always negative, it is possible to infer the sign of the delta and only the magnitude of the difference may be coded. Moreover, since the difference cannot be 0, the magnitude of the delta minus 1 may be coded instead. In some scenarios, when the bottom region is not very clear, a single threshold may be sufficient. Thus, the G-PCC encoder 200 and the G-PCC decoder 300 may be configured to code a flag to indicate whether a second threshold exists. The syntax modifications to GPS according to which the G-PCC encoder 200 and the G-PCC decoder 300 may be configured may be shown in Table 3 as follows: [Table 3]

[0117] In another example, threshold1 may be signaled all the time, and threshold0 may be signaled conditionally based on the value of a flag.

[0118] In another example, the midpoint (m) of the two thresholds may be signaled, and the distance (w) of the midpoint from either threshold may be signaled, and then the two thresholds may be derived as mw and m+w. These values ​​may be signaled using fixed-length or variable-length coding.

[0119] In another example, the G-PCC encoder 200 and the G-PCC decoder 300 may code the data for these thresholds at the GPS level, with the possibility to overwrite / refine the thresholds at the GDH level.

[0120] In another example, the G-PCC encoder 200 and the G-PCC decoder 300 may code these thresholds together with the global motion information (rotation and translation coefficients).

[0121] In another example, the G-PCC encoder 200 and the G-PCC decoder 300 may code the thresholds in a separate parameter set, such as a parameter set dedicated to motion-related parameters.

[0122] In some examples, in addition to or instead of the techniques discussed above, the G-PCC encoder 200 and the G-PCC decoder 300 may be configured to implicitly classify points, for example as object points or ground points, through coding points in a slice that correspond to a class. For example, if a point cloud includes an object point class and a ground (or road) point class, the G-PCC encoder 200 and the G-PCC decoder 300 may code an object slice that includes a first subset of points that are all classified as object points, and a ground or road slice that includes a second subset of points that are all classified as ground points or road points. More than two classes may be used in this manner. In general, the G-PCC encoder 200 and the G-PCC decoder 300 may be configured to determine that there is one slice for each class of points, and that all points in a given slice should be classified according to the corresponding class for the given slice. An explicit classification algorithm is not necessary in this example, which may reduce the computation to be performed by the G-PCC encoder 200 and the G-PCC decoder 300.

[0123] More generally, the G-PCC encoder 200 and the G-PCC decoder 300 may be configured to perform the following techniques, either alone or in any combination with various other techniques of this disclosure. 1. Classification (or Partitioning) of Points of a Point Cloud into M Groups The G-PCC encoder 200 and the G-PCC decoder 300 may be configured according to one of the techniques of this disclosure or other means to achieve classification of points into M groups. Examples of groups include roads, medians, nearby cars or vehicles, buildings, signs, traffic signals, pedestrians, etc. Note that each car / vehicle / building etc. may be classified as a separate group. b. A group may contain points that represent objects or that are spatially adjacent to one another. 2. The G-PCC encoder 200 and the G-PCC decoder 300 may specify N slice groups, where N<=M. The G-PCC encoder 200 and the G-PCC decoder 300 may associate each of the M groups with one of the N slice groups. The G-PCC encoder 200 and the G-PCC decoder 300 may jointly code points that belong to the slice groups. For example, a "Ground" slice group may include points belonging to the "Road" and "Median" groups, a "Static" slice group may include points belonging to "Buildings" and "Signs", and a "Dynamic" slice group may include groups such as cars / vehicles, or "Pedestrians". b. More generally, the G-PCC encoder 200 and the G-PCC decoder 300 may code one or more groups that share some property into a slice group. For example, groups that may have similar relative motion with respect to the LIDAR sensor / vehicle may be coded into one slice group. c. In another example, the G-PCC encoder 200 and the G-PCC decoder 300 may be configured to determine that each group of points having a certain property belongs to a separate slice group. d. A point of a group may be associated with more than one slice group (eg, a point may be repeated). 3. The G-PCC encoder 200 and the G-PCC decoder 300 may code points that belong to each slice group in one or more slices. 4. The G-PCC encoder 200 and the G-PCC decoder 300 may identify slices that belong to a slice group based on an index value (eg, slice index) or a label (slice type or slice group type). Each slice group may be associated with a slice type / slice group type, which may be signaled in each slice of the slice group. i. For example, an index / label of [0,N-1] may be associated with each slice group, and the G-PCC encoder 200 and the G-PCC decoder 300 may code index / label "i" in a slice that belongs to the i-th slice group (0<=i<=N-1). In another example, a point cloud may have two slice groups, S1 and S2, each of which may be coded as three slices, for a total of six slices. Each of the slices in S1 may have slice type 0, and each of the slices in S2 may have slice type 1. b. In another example, each slice may be associated with a slice number in the slice index, and slices that belong to a particular slice group may be identified using the slice number / index. For example, a point cloud may have two slice groups, S1 and S2, each of which may be coded as three slices, for a total of six slices. The slices in S1 may have slice numbers 0, 1, and 2, and the slices in S2 may have slice numbers 3, 4, and 5. c. In some examples, the slice identifier may be a combination of a slice group identifier / type and a slice number. i. For example, a point cloud may have two slice groups, S1 and S2, each of which may be coded as three slices, for a total of six slices. The slices in S1 may have identifiers (0,0), (0,1), (0,2), where the first number in each tuple is the slice type and the second number is the slice number within the slice group. Similarly, the slices in S2 may have identifiers (1,0), (1,1), (1,2). d. The slice identifier slice type, slice group type, and slice number may be signaled in the slice. 5. The G-PCC encoder 200 and the G-PCC decoder 300 may code data that references a slice for prediction. A slice may reference another slice for prediction. The reference slice may belong to the same picture (intra prediction) or another picture (inter prediction). a. The G-PCC encoder 200 and the G-PCC decoder 300 may identify the reference slices using one or more of the following: i. Reference frame number or frame counter ii. Reference slice identifier (slice type / group type, slice number, slice identifier, etc.) b. In some examples, the G-PCC encoder 200 and the G-PCC decoder 300 may be configured according to the constraint that a slice can only reference other slices that belong to the same slice type / slice group type. In this case, the reference slice type / slice group type does not need to be signaled. c. In another example, a slice may be allowed to reference all points belonging to a frame or slice group, in which case the reference slice number does not need to be signaled since all slices of the frame / slice group may be referenced for prediction. d. In another example, two or more slice identifiers may be signaled that identify multiple slices that may be referenced for prediction. 6. The G-PCC encoder 200 and the G-PCC decoder 300 may associate a first set of motion parameters for each point, which may be used to compensate the location of the point, and which may be used as a reference for prediction. In one example, the motion parameters associated with a point may be the motion parameters associated with the slice that contains the point. b. In one example, the motion parameters associated with a slice may be motion parameters associated with a slice group that includes the slice. c. In one example, the motion parameters associated with a slice group may be motion parameters associated with a frame that includes the slice group. d. Motion parameters may be signaled in parameter sets such as SPS, GPS, etc., slice headers, or other parts of the bitstream. e. Although the above description refers to motion parameters, this may apply to any set of motion parameters (eg, rotation matrices / parameters, translation vectors / parameters, etc.). In some examples, motion parameters used to apply motion compensation for points in a reference frame may be signaled in the current frame or in a frame that is not a reference frame. For example, if frame 1 uses points from frame 0 for prediction, the motion parameters applied to points in frame 0 may be signaled along with frame 1. g. In one example, a reference index for a slice / slice group of a reference frame may be signaled in the current frame (in a parameter set or slice or other syntax structure). i. In one example, one or more tuples (motion parameters, reference index) may be signaled along with the current frame (or slice), where the reference index identifies the point in the reference frame (slice / slice group / region) to which the respective motion parameters apply. h. In one example, the motion parameters may be a set of global motion parameters that apply to all points in a slice, slice group, region, or frame.

[0124] One or more of the techniques of this disclosure may also be applied to attributes, for example in addition to or instead of being applied to points.

[0125] In some examples, the G-PCC encoder 200 and the G-PCC decoder 300 may be configured to specify one or more regions in the point cloud. The G-PCC encoder 200 and the G-PCC decoder 300 may further associate motion parameters with each region. The G-PCC encoder 200 and the G-PCC decoder 300 may code data in the bitstream representing the motion parameters associated with the region. The G-PCC encoder 200 and the G-PCC decoder 300 may use the motion parameters to compensate the location of the point. The G-PCC encoder 200 and the G-PCC decoder 300 may use the compensated point as a reference / prediction for coding the location of the point in the current frame. In some cases, the use of a region (compared to a slice) for classification may achieve better compression performance because the G-PCC encoder 200 and the G-PCC decoder 300 may jointly code points that belong to different regions. 1. The G-PCC encoder 200 and the G-PCC decoder 300 may code data representing one or more regions in a point cloud. a. The G-PCC encoder 200 and the G-PCC decoder 300 may code data representing a value N representing the number of regions, as well as parameters specifying each of the N regions. i. In some examples, N may be constrained to be within a range of values ​​(eg, N may be constrained to be less than a certain value, such as 10). b. The G-PCC encoder 200 and the G-PCC decoder 300 may code parameters for each region in the bitstream. In some examples, a region may be specified using one or more of the following parameters: i. The upper and lower x, y, and z coordinates (or any other coordinate system used to code the point cloud) that define the region. ii. In some examples, one or more of the upper or lower bounds may not be specified, in which case the G-PCC encoder 200 and the G-PCC decoder 300 may use default values ​​appropriate for the coordinates and coordinate system as inferred values. 1. For example, in a spherical domain (r, phi, laserId), if the bounds of phi are not signaled, the upper and lower bounds may be inferred to correspond to 360 degrees and 0 degrees, respectively. 2. Motion parameters may be associated with each region. Motion compensation may be applied to one or more points belonging to the region to obtain a compensated location / point. The compensated location / point may be used as a reference for prediction of a point in the current point cloud frame. One or more methods disclosed in this disclosure for signaling motion parameters may be applied to signal the motion parameters of each region. For example, the G-PCC encoder 200 and the G-PCC decoder 300 may code the motion parameters for each region in a parameter set (e.g., SPS, GPS) or other parts of the bitstream (e.g., slice header, or separate syntax structure).

[0126] The G-PCC encoder 200 and the G-PCC decoder 300 may be configured to perform any of the various techniques of this disclosure in various combinations. For example, motion parameters for a reference frame may be specified with respect to a region, but one or more slice groups may be specified for the current frame. A slice group may be associated with a region (explicitly or implicitly), and reference points from the region may be used to predict points in the slice group. In another example, points in a region may be coded as a slice or slice group.

[0127] 12 is a graph 460 illustrating an example derivation of thresholds using a histogram in accordance with the techniques of this disclosure. The graph 460 represents an example histogram of collected heights (z values) of the point cloud data. The G-PCC encoder 200 may use the histogram to calculate z_bottom 462 and z_top 464.

[0128] In one example implementation, the G-PCC encoder 200 may downsample (subsample) the constellation using a size of hist_bin_size, which may be defined as: hist_bin_size = int((max_box_t - min_box_t) / hist_scale) where max_box_t and min_box_t are the ranges of z values ​​in the group that are used to obtain the thresholds: max_box_t may be lower than the maximum value of z in the group, and min_box_t may be higher than the minimum value of z in the group.

[0129] Next, the G-PCC encoder 200 may derive a histogram of points whose z-values ​​are in the range min_box_t to max_box_t as follows (this is example Python code, but other implementations in other languages ​​or hardware may also be used): n, bins = np.histogram(source_points_ori, hist_bin_size, (min_box_t, max_box_t)) In this example, np is a representation of the numpy library (numpy.org) and source_point_ori is the set of points whose z-values ​​are in the range min_box_t to max_box_t.

[0130] After this, the G-PCC encoder 200 may calculate the standard deviation (std) 466 of the histogram, for example according to the following Python code (although other implementations in other languages ​​or hardware may also be used): mids = 0.5*(bins[1:] + bins[:-1]) probs = n / np.sum(n) mean = np.sum(probs * mids) std = np.sqrt(np.sum(probs * (mids - mean)**2))

[0131] Finally, in this example, the G-PCC encoder 200 may derive z_bottom 462 and z_top 464 as follows: The G-PCC encoder 200 determines the bin index (top_idx_n, bin 470 in the example of FIG. 12) in the histogram where the count of points is the highest. The G-PCC encoder 200 determines the thresholds z_top (z_top 464) and z_bottom (z_bottom 462) by shifting right and left from bin 470 (i.e., the bin where the count of points is the highest) by a value related to std, e.g., 1*std 466 and 1.5*std 468 in the example of FIG. 12. The following Python code represents an example technique by which the thresholds may be derived: top_idx_n = np.where(n == n.max()) z_top = min(bins[top_idx_n] + w_top*std, max_box_t) z_bottom = max(bins[top_idx_n] - w_bottom*std, max_box_t) Here, w_top and w_bottom are predetermined positive values.

[0132] In the example of Figure 12, (max_box_t, min_box_t) are set equal to (-500, -4000) for the 100th frame of the collected data set. In Figure 12, (w_top, w_bottom) are set equal to (1, 1.5).

[0133] FIG. 13 is a conceptual diagram illustrating labeling points in a point cloud 470 as road points 474 and object points 472 according to the techniques of this disclosure. A car (not shown in FIG. 13) equipped with a LIDAR system located approximately at point 476 may collect data from the surrounding environment to construct the point cloud 470. A G-PCC encoder, such as G-PCC encoder 200 in the car, may determine a threshold for classifying points in the point cloud 470 into road points or object points. After determining the threshold (e.g., according to the techniques of FIG. 12), the G-PCC encoder 200 may label points in the point cloud 470 as ground / road points 474 or object points 472.

[0134] Figure 13 depicts a visualization of these sets of points, with object points 472 shaded darkly and road points 474 (also referred to as ground points) shaded lightly. As can be seen in the example of Figure 13, the lightly shaded road points 474 are spread across a generally flat surface (e.g., the ground or roads on the ground), while the darkly shaded object points 472 generally define objects such as fences, signs, buildings, or other objects near the location of the car when point cloud 470 was generated.

[0135] 14 is a flow chart illustrating an example method for encoding a point cloud in accordance with the techniques of this disclosure. The method of FIG. 14 is described with respect to the G-PCC encoder 200 of FIG. 1 and FIG. 2. Other G-PCC encoding devices may be configured to perform this method or a similar method.

[0136] First, the G-PCC encoder 200 may obtain a point cloud to be encoded, e.g., the current cloud 140 of FIG. 3. The point cloud may include a set of points, each of which has a geometric location (e.g., represented by (x, y, z) coordinates) and one or more attributes. The G-PCC encoder 200 may then determine height values ​​of the points in the point cloud (500), e.g., using the z values ​​of the points' geometric locations. The G-PCC encoder 200 may then determine top and bottom thresholds (502) and use the thresholds to classify ground points and object points in the point cloud (504). For example, the G-PCC encoder 200 may determine the thresholds using the techniques discussed above with respect to FIG. 11 and FIG. 12. The G-PCC encoder 200 may then classify points between the top and bottom thresholds as ground points and other points as object points. The G-PCC encoder 200 may also encode a data structure (e.g., SPS, GPS, GDH, etc.) that includes data representing the top and bottom thresholds. The data structure may follow any of the examples in Table 1-3 above.

[0137] The G-PCC encoder 200 may then calculate a global motion vector for the object points (506). For example, as shown in FIG. 3, the G-PCC encoder 200 may calculate a global motion vector 132 for the set of object points. The global motion vector may generally represent the motion vector that best produces the predicted constellation 134 (i.e., produces a predicted constellation that includes points that best match the current constellation 140 to the reference constellation 130). After obtaining the global motion vector, the G-PCC encoder 200 may use the global motion vector 132 for the reference constellation 130 to generate the predicted constellation 134 (508).

[0138] The G-PCC encoder 200 may then use the predicted set 134 to determine a context for encoding the occupancy of the nodes of the current set 140 by the determined body points (510). The G-PCC encoder 200 may further use the context to entropy encode data representing the occupancy of the nodes by the body points (512). Specifically, for a given node of the current set 140, the G-PCC encoder 200 may determine whether the corresponding node (having the same size and location in the predicted set 134) is occupied by at least one body point. If the corresponding node is occupied (i.e., includes at least one body point), the G-PCC encoder 200 may determine a context for encoding a value indicating whether the current node is occupied as having a high probability indicating that the current node of the current set 140 is also occupied. If the corresponding node is unoccupied (i.e., does not contain any object points), the G-PCC encoder 200 may determine the context for encoding a value indicating whether the current node is occupied as having a high probability of indicating that the current node in the current group 140 is unoccupied.

[0139] The G-PCC encoder 200 may then encode the value using the determined context. If the current node is unoccupied, the G-PCC encoder 200 may proceed to a new node. On the other hand, if the current node is occupied, the G-PCC encoder 200 may partition the current node into eight partial nodes and encode the occupancy data for each of the eight partial nodes in the same manner.

[0140] The G-PCC encoder 200 may then also separately encode data representing the occupancy of the nodes by the ground points (514). For example, the G-PCC encoder 200 may encode the data representing the occupancy of the nodes by the ground points using a second, different global motion vector, local motion vector, and / or intra prediction.

[0141] Thus, the method of FIG. 14 represents an example of a method for coding point cloud data, which includes the steps of determining elevation values ​​for points in the point cloud, classifying the points into a set of ground points or a set of object points according to the elevation values, and coding the ground points and object points according to the classification.

[0142] 15 is a flowchart illustrating an example method for decoding a point cloud according to the techniques of this disclosure. The method of FIG. 15 is described as being performed by the G-PCC decoder 300 of FIG. 1 and FIG. 4. However, in other examples, other decoding devices may be configured to perform this method or a similar method.

[0143] First, the G-PCC decoder 300 may determine the top and bottom thresholds (520). For example, the G-PCC decoder 300 may encode a data structure (e.g., SPS, GPS, GDH, etc.) that includes data representing the top and bottom thresholds. The data structure may follow any of the examples in Table 1-3 above. The G-PCC decoder 300 may further decode global motion vectors for object points, i.e., points within nodes that have height values ​​outside the range between the top and bottom thresholds (522). For example, the G-PCC decoder 300 may decode data representing the occupancy of nodes above the top threshold and / or below the bottom threshold using the global motion vectors, as follows:

[0144] The G-PCC decoder 300 may use a global motion vector (e.g., global motion vector 152) relative to the reference group 150 to form a predicted group (e.g., predicted group 154 ​​of FIG. 5) (524). The G-PCC decoder 300 may then use the points in the predicted group 154 ​​to determine a context for decoding data representing the occupancy of a node in the current group 160 (526). Specifically, for a given node of the current group 160, the G-PCC decoder 300 may determine whether the corresponding node (having the same size and location in the predicted group 154) is occupied by at least one object point. If the corresponding node is occupied (i.e., includes at least one object point), the G-PCC decoder 300 may determine a context for decoding a value indicating whether the current node is occupied, as having a high probability indicating that the current node of the current group 160 is also occupied. If the corresponding node is unoccupied (i.e., does not contain any object points), the G-PCC decoder 300 may determine the context for encoding a value indicating whether the current node is occupied as having a high probability of indicating that the current node in the current group 160 is unoccupied.

[0145] The G-PCC decoder 300 may then use the context to entropy decode the data representing the occupancy of the node (528). If the decoded data indicates that the current node was unoccupied, the G-PCC decoder 300 may proceed to a new node. On the other hand, if the current node was occupied, the G-PCC decoder 300 may partition the current node into eight partial nodes and decode the occupancy data for each of the eight partial nodes in the same manner.

[0146] The G-PCC decoder 300 may also separately decode data representing occupancy of nodes by ground points (530), eg, using different global motion vectors, local motion vectors, and / or intra prediction.

[0147] 16 is a conceptual diagram illustrating a laser package 600, such as a LIDAR sensor or other system including one or more lasers, scanning a point in three-dimensional space. Laser package 600 may correspond to LIDAR 380 of FIG. 7. Data source 104 (FIG. 1) may include laser package 600.

[0148] As shown in FIG. 16, a laser package 600 can be used to capture a point cloud, i.e., a sensor scans points in 3D space. However, it should be understood that some point clouds may not be generated by an actual LIDAR sensor, but may be encoded as if they were. In the example of FIG. 16, the laser package 600 includes a LIDAR head 602 that includes multiple lasers 604A-604E (collectively "lasers 604") aligned in a perpendicular plane at different angles to an origin. The laser package 600 may rotate about a vertical axis 608. The laser package 600 may use the returned laser light to determine distances and locations of points in the point cloud. The laser beams 606A-606E (collectively "laser beams 606") emitted by the lasers 604 of the laser package 600 may be characterized by a set of parameters. Distances indicated by arrows 610, 612 indicate exemplary laser correction values ​​for lasers 604B, 604A, respectively.

[0149] Some lasers 604 may generally identify object points, while other lasers 604 may generally identify ground points. Using the techniques of this disclosure, points may be classified as either ground points or object points and encoded or decoded accordingly.

[0150] FIG. 17 is a conceptual diagram illustrating an example distance measurement system 900 that may be used with one or more techniques of the present disclosure. In the example of FIG. 17, the distance measurement system 900 includes an illuminator 902 and a sensor 904. The illuminator 902 may emit light 906. In some examples, the illuminator 902 may emit the light 906 as one or more laser beams. The light 906 may be one or more wavelengths, such as infrared wavelengths or visible light wavelengths. In other examples, the light 906 is not a coherent laser light. When the light 906 strikes an object, such as an object 908, the light 906 produces returning light 910. The returning light 910 may include backscattered light and / or reflected light. The returning light 910 may pass through a lens 911 that directs the returning light 910 to produce an image 912 of the object 908 on the sensor 904. The sensor 904 generates a signal 914 based on the image 912. Image 912 may comprise a set of points (eg, as represented as dots in image 912 of FIG. 17).

[0151] In some examples, the illuminator 902 and sensor 904 may be mounted on a rotating structure such that the illuminator 902 and sensor 904 capture a 360-degree view of the environment. In other examples, the distance measurement system 900 may include one or more optical components (e.g., mirrors, collimators, diffraction gratings, etc.) that enable the illuminator 902 and sensor 904 to detect objects within a certain range (e.g., up to 360 degrees). Although the example of FIG. 17 shows only a single illuminator 902 and sensor 904, the distance measurement system 900 may include multiple sets of illuminators and sensors.

[0152] In some examples, the illuminator 902 generates a structured light pattern. In such examples, the distance measurement system 900 may include a plurality of sensors 904 on which respective images of the structured light pattern are formed. The distance measurement system 900 may use the difference between the images of the structured light pattern to determine the distance to an object 908 from which the structured light pattern is backscattered. The structured light based distance measurement system may have a high level of accuracy (e.g., accuracy in the sub-millimeter range) when the object 908 is relatively close to the sensor 904 (e.g., 0.2 meters to 2 meters). This high level of accuracy may be useful in facial recognition applications, such as for unlocking mobile devices (e.g., cell phones, tablet computers, etc.) and security applications.

[0153] In some examples, the distance measurement system 900 is a time of flight (ToF) based system. In some examples where the distance measurement system 900 is a ToF based system, the illuminator 902 generates a pulse of light. In other words, the illuminator 902 may modulate the amplitude of the emitted light 906. In such examples, the sensor 904 detects the returning light 910 from the pulse of light 906 generated by the illuminator 902. The distance measurement system 900 may then determine the distance to the object 908 from which the light 906 is backscattered based on the delay between when the light 906 is emitted and when it is detected, and the known speed of light in air. In some examples, rather than (or in addition to) modulating the amplitude of the emitted light 906, the illuminator 902 may modulate the phase of the emitted light 906. In such an example, the sensor 904 may detect the phase of the returning light 910 from the object 908 and determine the distance to a point on the object 908 using the speed of light and based on the time difference between when the illuminator 902 generates the light 906 with a particular phase and when the sensor 904 detects the returning light 910 with a particular phase.

[0154] In other examples, the point cloud may be generated without the use of the illuminator 902. For example, in some examples, the sensor 904 of the distance measurement system 900 may include two or more optical cameras. In such examples, the distance measurement system 900 may use the optical cameras to capture a stereo image of an environment, including the object 908. The distance measurement system 900 (e.g., the point cloud generator 920) may then calculate the difference between the positions in the stereo image. The distance measurement system 900 may then use the difference to determine the distance to the positions shown in the stereo image. From these distances, the point cloud generator 920 may generate the point cloud.

[0155] The sensor 904 may also detect other attributes of the object 908, such as color and reflectance information. In the example of Figure 17, a point cloud generator 920 may generate a point cloud based on the signal 918 generated by the sensor 904. The distance measurement system 900 and / or the point cloud generator 920 may form part of the data source 104 (Figure 1).

[0156] FIG. 18 is a conceptual diagram illustrating an example vehicle-based scenario in which one or more techniques of the present disclosure may be used. In the example of FIG. 18, the vehicle 1000 includes a laser package 1002, such as a LIDAR system. The laser package 1002 may be implemented in the same manner as the laser package 600 (FIG. 16). Although not shown in the example of FIG. 18, the vehicle 1000 may also include a data source, such as the data source 104 (FIG. 1), and a G-PCC encoder, such as the G-PCC encoder 200 (FIG. 1). In the example of FIG. 18, the laser package 1002 emits a laser beam 1004 that reflects off a pedestrian 1006 or other object on the road. The data source of the vehicle 1000 may generate a point cloud based on a signal generated by the laser package 1002. The G-PCC encoder of the vehicle 1000 may encode the point cloud to generate a bit stream 1008, such as the geometry bit stream of FIG. 2 and the attribute bit stream of FIG. 2. The bitstream 1008 may include many fewer bits than the unencoded point cloud obtained by the G-PCC encoder. An output interface of the vehicle 1000 (e.g., output interface 108 (FIG. 1)) may transmit the bitstream 1008 to one or more other devices. Thus, the vehicle 1000 may be able to transmit the bitstream 1008 to other devices more quickly than unencoded point cloud data. In addition, the bitstream 1008 may require less data storage capacity.

[0157] Techniques of this disclosure may further reduce the number of bits in the bitstream 1008. For example, encoding object points separately from ground points, e.g., using global motion information for the object points, may reduce the number of bits in the bitstream 1008 associated with the object points.

[0158] In the example of FIG. 18, the vehicle 1000 may transmit the bit stream 1008 to another vehicle 1010. The vehicle 1010 may include a G-PCC decoder, such as the G-PCC decoder 300 (FIG. 1). The G-PCC decoder of the vehicle 1010 may decode the bit stream 1008 to reconstruct a point cloud. The vehicle 1010 may use the reconstructed point cloud for various purposes. For example, the vehicle 1010 may determine that a pedestrian 1006 is on the road in front of the vehicle 1000 based on the reconstructed point cloud and therefore begin to slow down, for example, even before the driver of the vehicle 1010 recognizes that the pedestrian 1006 is on the road. Thus, in some examples, the vehicle 1010 may perform an autonomous navigation operation, generate a notification or alert, or perform another activity based on the reconstructed point cloud.

[0159] Additionally or alternatively, the vehicle 1000 may transmit the bitstream 1008 to the server system 1012. The server system 1012 may use the bitstream 1008 for various purposes. For example, the server system 1012 may store the bitstream 1008 for later reconstruction of a point cloud. In this example, the server system 1012 may use the point cloud along with other data (e.g., vehicle telemetry data generated by the vehicle 1000) to train an autonomous driving system. In other examples, the server system 1012 may store the bitstream 1008 for later reconstruction for forensic accident investigation (e.g., if the vehicle 1000 collides with a pedestrian 1006).

[0160] FIG. 19 is a conceptual diagram of an example extended reality system in which one or more techniques of the present disclosure may be used. Extended reality (XR) is a term used to encompass a wide range of technologies, including augmented reality (AR), mixed reality (MR), and virtual reality (VR). In the example of FIG. 19, a first user 1100 is located at a first location 1102. The user 1100 is wearing an XR headset 1104. Instead of the XR headset 1104, the user 1100 may use a mobile device (e.g., a mobile phone, a tablet computer, etc.). The XR headset 1104 includes a depth detection sensor, such as a LIDAR system, that detects the location of a point on an object 1106 at the location 1102. A data source of the XR headset 1104 may use a signal generated by the depth detection sensor to generate a point cloud representation of the object 1106 at the location 1102. The XR headset 1104 may include a G-PCC encoder (e.g., the G-PCC encoder 200 of FIG. 1) configured to encode the point cloud to generate a bitstream 1108.

[0161] Techniques of this disclosure may further reduce the number of bits in the bitstream 1108. For example, encoding the object points separately from the ground points, e.g., using common global motion information for the object points, may reduce the number of bits in the bitstream 1108 associated with the third laser angle.

[0162] The XR headset 1104 may transmit the bitstream 1108 (e.g., over a network such as the Internet) to an XR headset 1110 worn by a user 1112 at a second location 1114. The XR headset 1110 may decode the bitstream 1108 to reconstruct the point cloud. The XR headset 1110 may use the point cloud to generate an XR visualization (e.g., an AR, MR, VR visualization) representing the object 1106 at the location 1102. Thus, in some examples, such as when the XR headset 1110 generates a VR visualization, the user 1112 at the location 1114 may experience a 3D immersive feeling of the location 1102. In some examples, the XR headset 1110 may determine the location of the virtual object based on the reconstructed point cloud. For example, the XR headset 1110 may determine, based on the reconstructed point cloud, that the environment (e.g., location 1102) includes a flat surface and determine that a virtual object (e.g., a cartoon character) should be placed on the flat surface. The XR headset 1110 may generate an XR visualization with the virtual object at the determined location. For example, the XR headset 1110 may show the cartoon character sitting on the flat surface.

[0163] FIG. 20 is a conceptual diagram illustrating an example mobile device system in which one or more techniques of the present disclosure may be used. In the example of FIG. 20, a mobile device 1200, such as a mobile phone or tablet computer, includes a depth-detecting sensor, such as a LIDAR system, that detects the location of points on an object 1202 in an environment of the mobile device 1200. A data source of the mobile device 1200 may use signals generated by the depth-detecting sensor to generate a point cloud representation of the object 1202. The mobile device 1200 may include a G-PCC encoder (e.g., the G-PCC encoder 200 of FIG. 1) configured to encode the point cloud to generate a bitstream 1204. In the example of FIG. 20, the mobile device 1200 may transmit the bitstream to a remote device 1206, such as a server system or another mobile device. The remote device 1206 may decode the bitstream 1204 to reconstruct the point cloud. The remote device 1206 may use the point cloud for various purposes. For example, the remote device 1206 may generate a map of the environment of the mobile device 1200 using the point cloud. For example, the remote device 1206 may generate a map of the interior of a building based on the reconstructed point cloud. In another example, the remote device 1206 may generate imagery (e.g., computer graphics) based on the point cloud. For example, the remote device 1206 may use the points of the point cloud as vertices of a polygon and use color attributes of the points as a basis for shading the polygon. In some examples, the remote device 1206 may perform facial recognition using the point cloud.

[0164] The following clauses represent various examples of the techniques described in this disclosure.

[0165] Clause 1: A method for coding point cloud data, the method comprising the steps of determining elevation values ​​of points in the point cloud, classifying the points into a set of ground points or a set of object points according to the elevation values, and coding the ground points and the object points according to the classification.

[0166] Clause 2: The method of clause 1, wherein the step of classifying the points comprises the steps of determining a top threshold and a bottom threshold, classifying points having height values ​​between the top threshold and the bottom threshold into a set of ground points, and classifying points having height values ​​above the top threshold or below the bottom threshold into a set of object points.

[0167] Clause 3: Range of the i-th value {(x_min i , x_max i ), (y_min i , y_max i ), (z_min i , z_max i )}'s z_max i With the top threshold, z_min i 3. The method of any of clauses 1 and 2, wherein the bottom threshold comprises:

[0168] Clause 4: The method of clause 3, wherein the i-th value range comprises the i-th value range of the N value ranges.

[0169] Clause 5: x_min i and y_min i has a value of negative infinity, and x_max i and y_max i 5. The method of any of clauses 3 and 4, wherein has a value of infinity.

[0170] Clause 6: Any of the methods of clauses 2 to 5, wherein the step of coding the ground points and the object points further comprises the steps of quantizing the ground points and the object points by a scaling factor, and quantizing the top and bottom thresholds by the scaling factor.

[0171] Clause 7: Any of the methods of clauses 2 to 6, wherein the step of coding the object points comprises the steps of deriving a set of global motions for the object points and predicting the object points using the set of global motions.

[0172] Clause 8: The method of clause 7, wherein deriving a set of global motions comprises deriving the set of global motions from only the points of the object.

[0173] Clause 9: The method of any of clauses 7 and 8, wherein the set of global motions comprises a first set of global motions, and coding the ground points comprises deriving a second set of global motions for the ground points, and predicting the ground points using the second set of global motions.

[0174] Clause 10: The method of clause 9, wherein deriving the second set of global motions comprises deriving the second set of global motions from only ground points.

[0175] Clause 11: Any of the methods of clauses 7 to 10, wherein the step of deriving a set of global motions comprises the step of deriving a rotation matrix and a translation vector, and the step of coding the object points comprises the step of applying the rotation matrix and the translation vector to reference points of a reference frame.

[0176] Clause 12: The method of clause 11, wherein the step of coding the object points further comprises the steps of determining local node motion vectors for nodes of the predictive tree, the nodes including respective sets of reference points of a reference frame, and applying the local node motion vectors to the nodes.

[0177] Clause 13: The method of any of clauses 2 to 12, wherein determining the top and bottom thresholds comprises determining top and bottom thresholds for a group of pictures (GOP) including a plurality of frames including the point cloud.

[0178] Clause 14: Any of the methods of clauses 2 to 12, wherein the step of determining the top and bottom thresholds comprises determining top and bottom thresholds for a sequence parameter set (SPS) corresponding to a plurality of frames including the point cloud.

[0179] Clause 15: The method of any of clauses 13 and 14, wherein determining the top and bottom thresholds comprises determining top and bottom thresholds for a first frame of the plurality of frames.

[0180] Clause 16: The method of any of clauses 13 and 14, wherein determining the top and bottom thresholds comprises determining the top and bottom thresholds as a weighted average of thresholds for a plurality of frames.

[0181] Clause 17: The method of any of clauses 2 to 16, further comprising the step of coding a global parameter set (GPS) including data representative of at least one of the apex threshold or the bottom threshold.

[0182] Clause 18: The method of clause 17, wherein the step of coding the GPS comprises the step of coding a value for a top threshold and a flag indicating whether the data should be coded for a bottom threshold.

[0183] Clause 19: A method according to any of clauses 17 and 18, wherein the step of coding data for at least one of the top threshold and the bottom threshold comprises the step of coding a value of geom_globmotion_threshold0 representing the top threshold and the step of coding a value of geom_globmotion_threshold1 representing the bottom threshold.

[0184] Clause 20: The method of any of clauses 17 to 19, wherein the step of coding the data representing at least one of the top threshold or the bottom threshold comprises the step of coding the data representing at least one of the top threshold or the bottom threshold using a respective unsigned integer zeroth order exponential Golomb value.

[0185] Clause 21: The method of any of clauses 17 to 19, wherein the step of coding the data representing at least one of the top threshold or the bottom threshold comprises the step of coding the data representing at least one of the top threshold or the bottom threshold using a respective signed integer zeroth order exponential-Golomb value.

[0186] Clause 22: The method of any of clauses 17 to 19, wherein the step of coding data representing at least one of the top or bottom thresholds comprises the step of coding the data representing at least one of the top or bottom thresholds using a respective signed fixed length value, and the method further comprises the step of coding data representing a number of bits allocated to at least one of the top or bottom thresholds.

[0187] Clause 23: The method of any of clauses 17 to 22, wherein the step of coding data representing at least one of the top threshold or the bottom threshold comprises the step of coding data representing a midpoint between the top threshold and the bottom threshold, and coding data representing a distance from the midpoint to the top and bottom thresholds.

[0188] Clause 24: The method of any of clauses 17 to 23, further comprising the step of coding a geometry data unit header (GDH) including data that overwrites or refines the GPS data for at least one of the top threshold or the bottom threshold.

[0189] Clause 25: Any of the methods of clauses 2 to 24, wherein the step of determining the top and bottom thresholds comprises the steps of determining a maximum histogram height value, max_box_t, determining a minimum histogram height value, min_box_t, determining a histogram scale value, hist_scale, determining a histogram bin size value, hist_bin_size, according to int((max_box_t - min_box_t) / hist_scale), generating a histogram of points with height values ​​in the range from min_box_t to max_box_t, determining the bin in the histogram which has the maximum number of height values, and determining the top and bottom thresholds according to an offset from the bin with the maximum number of height values, the offset being defined according to a respective multiple of the standard deviation.

[0190] Clause 26: A method of coding point cloud data, the method comprising the steps of determining a first class associated with a first slice of a frame of point cloud data, the first slice including a first one or more points, determining that the first one or more points correspond to the first class, and coding the first one or more points in accordance with the determination that the first one or more points correspond to the first class; determining a second class associated with a second slice of the frame of point cloud data, the second slice including a second one or more points, determining that the second one or more points correspond to the second class, and coding the second one or more points in accordance with the determination that the second one or more points correspond to the second class.

[0191] Clause 27: A method comprising a combination of any of the methods of clauses 1 to 25 and clause 26.

[0192] Clause 28: The method of any of clauses 26 and 27, further comprising the steps of determining a third class associated with a third slice of the frame of point cloud data, the third slice including a third one or more points; determining that the third one or more points correspond to the third class; and coding the third one or more points in accordance with the determination that the third one or more points correspond to the third class.

[0193] Clause 29: Any of the methods of clauses 26 to 28, wherein the first class and the second class comprise at least one of a road, a median strip, a nearby vehicle, a building, a sign, a traffic signal, or a pedestrian.

[0194] Clause 30: The method of any of clauses 26 to 29, further comprising the step of coding data representing a slice group for a first slice, wherein coding the first one or more points comprises coding the first one or more points together with other points included in one or more other slices corresponding to the slice group.

[0195] Clause 31: The method of clause 30, further comprising the steps of determining a third class associated with a third slice of the frame of point cloud data, the third slice being one of the one or more other slices corresponding to the slice group, and coding a third one or more points of the third slice together with the first one or more points.

[0196] Clause 32: The method of any of clauses 30 and 31, further comprising coding an index value for each of the slices representing a corresponding slice group.

[0197] Clause 33: The method of any of clauses 26 to 32, wherein the step of coding the first one or more points comprises predicting at least one of the first one or more points from a third one or more points of a third slice.

[0198] Clause 34: The method of clause 33, wherein the frame comprises a first frame and the third slice forms part of a second frame distinct from the first frame.

[0199] Clause 35: The method of any of clauses 26 to 34, further comprising determining a respective motion parameter for each of the first one or more points and the second one or more points.

[0200] Clause 36: A method of coding point cloud data, comprising the steps of determining one or more regions of a frame of point cloud data, for each of the regions coding data representative of respective motion parameters for the region, and coding points of the region using the respective motion parameters for the region.

[0201] Clause 37: A method comprising a combination of any of the methods of clauses 1 to 35 and the method of clause 36.

[0202] Clause 38: The method of any of clauses 36 and 37, further comprising the step of coding data representative of the number of regions contained in the frame.

[0203] Clause 39: The method of any of clauses 36 to 38, further comprising the step of coding parameters specifying each of the regions of the frame.

[0204] Clause 40: The method of clause 39, wherein the parameters include at least one of an upper or lower bound on one or more of an x-coordinate of the region, a y-coordinate of the region, or a z-coordinate of the region.

[0205] Clause 41: The method of any of clauses 39 and 40, further comprising determining default values ​​for one or more coordinates of the region.

[0206] Clause 42: The method of any of clauses 36 to 41, wherein coding the points of the region comprises applying motion compensation to the points of the region using respective motion parameters.

[0207] Clause 43: A method for coding point cloud data, the method comprising the steps of determining elevation values ​​of points in the point cloud, classifying the points into a set of ground points or a set of object points according to the elevation values, and coding the ground points and the object points according to the classification.

[0208] Clause 44: The method of clause 43, wherein the step of classifying the points comprises the steps of determining a top threshold and a bottom threshold, classifying points having height values ​​between the top threshold and the bottom threshold into a set of ground points, and classifying points having height values ​​above the top threshold or below the bottom threshold into a set of object points.

[0209] Clause 45: Range of the i-th value {(x_min i , x_max i ), (y_min i , y_max i ), (z_min i , z_max i )}'s z_max i With the top threshold, z_min i 44. The method of claim 43, wherein the bottom threshold comprises:

[0210] Clause 46: The method of clause 45, wherein the i-th value range comprises the i-th value range of the N value ranges.

[0211] Article 47: x_min i and y_min i has a value of negative infinity, and x_max i and y_max i The method of clause 45, wherein has a value of infinity.

[0212] Clause 48: The method of clause 44, wherein the step of coding the ground points and the object points further comprises the steps of quantizing the ground points and the object points by a scaling factor, and quantizing the top and bottom thresholds by the scaling factor.

[0213] Clause 49: The method of clause 44, wherein coding the object points comprises deriving a set of global motions for the object points and predicting the object points using the set of global motions.

[0214] Clause 50: The method of clause 49, wherein deriving a set of global motions comprises deriving the set of global motions from only the points of the object.

[0215] Clause 51: The method of clause 50, wherein the set of global motions comprises a first set of global motions, and coding the ground points comprises deriving a second set of global motions for the ground points, and predicting the ground points using the second set of global motions.

[0216] Clause 52: The method of clause 51, wherein deriving the second set of global motions comprises deriving the second set of global motions from only the ground points.

[0217] Clause 53: The method of clause 49, wherein the step of deriving a set of global motions comprises the step of deriving a rotation matrix and a translation vector, and the step of coding the object points comprises the step of applying the rotation matrix and the translation vector to reference points of a reference frame.

[0218] Clause 54: The method of clause 53, wherein the step of coding the object points further comprises the steps of determining local node motion vectors for nodes of the predictive tree, the nodes including respective sets of reference points of a reference frame, and applying the local node motion vectors to the nodes.

[0219] Clause 55: The method of clause 44, wherein determining the top and bottom thresholds comprises determining top and bottom thresholds for a group of pictures (GOP) including a plurality of frames including the point cloud.

[0220] Clause 56: The method of clause 44, wherein determining the top and bottom thresholds comprises determining top and bottom thresholds for a sequence parameter set (SPS) corresponding to a plurality of frames including the point cloud.

[0221] Clause 57: The method of clause 56, wherein determining the top and bottom thresholds comprises determining top and bottom thresholds for a first frame of the plurality of frames.

[0222] Clause 58: The method of clause 56, wherein determining the top and bottom thresholds comprises determining the top and bottom thresholds as a weighted average of thresholds for a plurality of frames.

[0223] Clause 59: The method of clause 44, further comprising the step of coding a global parameter set (GPS) including data representative of at least one of the apex threshold or the bottom threshold.

[0224] Clause 60: The method of clause 59, wherein the step of coding the GPS comprises the step of coding a value for a top threshold and a flag indicating whether the data should be coded for a bottom threshold.

[0225] Clause 61: The method of clause 59, wherein the step of coding data for at least one of the top threshold and the bottom threshold comprises the step of coding a value of geom_globmotion_threshold0 representing the top threshold and coding a value of geom_globmotion_threshold1 representing the bottom threshold.

[0226] Clause 62: The method of clause 59, wherein coding the data representing at least one of the top threshold or the bottom threshold comprises coding the data representing at least one of the top threshold or the bottom threshold using a respective unsigned integer zeroth order exponential-Golomb value.

[0227] Clause 63: The method of clause 59, wherein coding the data representing at least one of the top threshold or the bottom threshold comprises coding the data representing at least one of the top threshold or the bottom threshold using a respective signed integer zeroth order exponential-Golomb value.

[0228] Clause 64: The method of clause 59, wherein the step of coding data representative of at least one of the top or bottom thresholds comprises the step of coding the data representative of at least one of the top or bottom thresholds using a respective signed fixed length value, and the method further comprises the step of coding data representative of a number of bits allocated to at least one of the top or bottom thresholds.

[0229] Clause 65: The method of clause 59, wherein the step of coding data representative of at least one of the top threshold or the bottom threshold comprises the steps of coding data representative of a midpoint between the top and bottom thresholds, and coding data representative of distances from the midpoint to the top and bottom thresholds.

[0230] Clause 66: The method of clause 59, further comprising the step of coding a geometry data unit header (GDH) including data that overwrites or refines the GPS data for at least one of the top threshold or the bottom threshold.

[0231] Clause 67: The method of clause 44, wherein the steps of determining the top and bottom thresholds comprise the steps of determining a maximum histogram height value, max_box_t, determining a minimum histogram height value, min_box_t, determining a histogram scale value, hist_scale, determining a histogram bin size value, hist_bin_size, according to int((max_box_t - min_box_t) / hist_scale), generating a histogram of points with height values ​​in the range from min_box_t to max_box_t, determining the bin in the histogram which has the maximum number of height values, and determining the top and bottom thresholds according to an offset from the bin with the maximum number of height values, the offset being defined according to a respective multiple of the standard deviation.

[0232] Clause 68: A method of coding point cloud data, the method comprising the steps of determining a first class associated with a first slice of a frame of point cloud data, the first slice including a first one or more points, determining that the first one or more points correspond to the first class, and coding the first one or more points in accordance with the determination that the first one or more points correspond to the first class; determining a second class associated with a second slice of the frame of point cloud data, the second slice including one or more points, determining that the second one or more points correspond to the second class, and coding the second one or more points in accordance with the determination that the second one or more points correspond to the second class.

[0233] Clause 69: The method of clause 68, further comprising the steps of determining a third class associated with a third slice of the frame of point cloud data, the third slice including a third one or more points; determining that the third one or more points correspond to the third class; and coding the third one or more points in accordance with the determination that the third one or more points correspond to the third class.

[0234] Clause 70: The method of clause 68, wherein the first class and the second class comprise at least one of a roadway, a median strip, a nearby vehicle, a building, a sign, a traffic signal, or a pedestrian.

[0235] Clause 71: The method of clause 68, further comprising the step of coding data representing a slice group for the first slice, wherein coding the first one or more points comprises coding the first one or more points together with other points included in one or more other slices corresponding to the slice group.

[0236] Clause 72: The method of clause 71, further comprising the steps of determining a third class associated with a third slice of the frame of point cloud data, the third slice being one of the one or more other slices corresponding to the slice group, and coding a third one or more points of the third slice together with the first one or more points.

[0237] Clause 73: The method of clause 71, further comprising the step of coding an index value for each of the slices representing a corresponding slice group.

[0238] Clause 74: The method of clause 68, wherein coding the first one or more points comprises predicting at least one of the first one or more points from a third one or more points of a third slice.

[0239] Clause 75: The method of clause 74, wherein the frame comprises a first frame and the third slice forms part of a second frame distinct from the first frame.

[0240] Clause 76: The method of clause 68, further comprising determining a respective motion parameter for each of the first one or more points and the second one or more points.

[0241] Clause 77: A method of coding point cloud data, the method comprising the steps of determining one or more regions of a frame of point cloud data, for each of the regions coding data representative of respective motion parameters for the region, and coding points of the region using the respective motion parameters for the region.

[0242] Clause 78: The method of clause 77, further comprising the step of coding data representative of the number of regions contained in the frame.

[0243] Clause 79: The method of clause 77, further comprising the step of coding parameters specifying each of the regions of the frame.

[0244] Clause 80: The method of clause 79, wherein the parameters include at least one of an upper or lower bound on one or more of an x-coordinate of the region, a y-coordinate of the region, or a z-coordinate of the region.

[0245] Clause 81: The method of clause 79, further comprising determining default values ​​for one or more coordinates of the region.

[0246] Clause 82: The method of clause 77, wherein coding the points of the region comprises applying motion compensation to the points of the region using respective motion parameters.

[0247] Clause 83: Any of the methods of clauses 1 to 82, wherein the coding step comprises a decoding step.

[0248] Clause 84: Any of the methods of clauses 1 to 83, wherein the coding step comprises the step of encoding.

[0249] Clause 85: A device for decoding point cloud data, the device comprising one or more means for performing the methods of any of clauses 1 to 84.

[0250] Clause 86: The device of clause 85, further comprising a display configured to display the point cloud data.

[0251] Clause 87: The device of any of clauses 85 and 86, wherein the device comprises one or more of a camera, a computer, a mobile device, a broadcast receiver device, or a set-top box.

[0252] Clause 88: The device of clauses 85 to 87, further comprising a memory configured to store the point cloud data.

[0253] Clause 89: A computer-readable storage medium having stored thereon instructions that, when executed, cause a processor to perform any of the methods of clauses 1 to 84.

[0254] Clause 90: A device for coding point cloud data, comprising means for determining height values ​​of points in the point cloud, means for classifying the points into a set of ground points or a set of object points according to the height values, and means for coding the ground points and the object points according to the classification.

[0255] Clause 91: The device of clause 90, wherein the means for classifying points comprises means for determining a top threshold and a bottom threshold, means for classifying points having height values ​​between the top threshold and the bottom threshold into a set of ground points, and means for classifying points having height values ​​above the top threshold or below the bottom threshold into a set of object points.

[0256] Clause 92: A device for coding point cloud data, the device comprising: means for determining a first class associated with a first slice of a frame of point cloud data, the first slice including a first one or more points; means for determining that the first one or more points correspond to the first class; and means for coding the first one or more points in accordance with a determination that the first one or more points correspond to the first class; means for determining a second class associated with a second slice of the frame of point cloud data, the second slice including a second one or more points; means for determining that the second one or more points correspond to the second class; and means for coding the second one or more points in accordance with a determination that the second one or more points correspond to the second class.

[0257] Clause 93: A device for coding point cloud data, comprising: means for determining one or more regions of a frame of point cloud data, means for coding data representing respective motion parameters for each of the regions, and means for coding each point of the regions using the respective motion parameters for the region containing the point.

[0258] Clause 94: A method for coding point cloud data, the method comprising the steps of determining elevation values ​​of points in the point cloud, classifying the points into a set of ground points or a set of object points according to the elevation values, and coding the ground points and the object points according to the classification.

[0259] Clause 95: The method of clause 94, wherein coding the object points comprises deriving a set of global motion information for the object points and predicting the object points using the set of global motion information.

[0260] Clause 96: The method of clause 95, wherein deriving a set of global motion information comprises deriving a set of global motion information for only the object points.

[0261] Clause 97: The method of clause 95, wherein the set of global motion information comprises a first set of global motion information, and coding the ground points comprises deriving a second set of global motion information for the ground points, and predicting the ground points using the second set of global motion information.

[0262] Clause 98: The method of clause 97, wherein deriving a second set of global motion information comprises deriving a second set of global motion information for only the ground points.

[0263] Clause 99: The method of clause 95, wherein the step of deriving a set of global motion information comprises the step of deriving a rotation matrix and a translation vector, and the step of coding the object points comprises the step of applying the rotation matrix and the translation vector to reference points of the reference frame.

[0264] Clause 100: The method of clause 99, wherein the step of coding the object points further comprises the steps of determining local node motion vectors for nodes of the predictive tree, the nodes including respective sets of reference points of a reference frame, and applying the local node motion vectors to the nodes.

[0265] Clause 101: The method of clause 94, wherein the step of classifying the points comprises the steps of determining a top threshold and a bottom threshold, classifying points having height values ​​between the top threshold and the bottom threshold into a set of ground points, and classifying points having height values ​​above the top threshold or below the bottom threshold into a set of object points.

[0266] Clause 102: The method of clause 101, wherein the top threshold comprises z_maxi and the bottom threshold comprises z_mini in the range {(x_mini, x_maxi), (y_mini, y_maxi), (z_mini, z_maxi)} of the i-th value.

[0267] Clause 103: The method of clause 102, wherein the i-th value range comprises an i-th value range of the N value ranges.

[0268] Clause 104: The method of clause 102, wherein x_mini and y_mini have a value of negative infinity, and x_maxi and y_maxi have a value of infinity.

[0269] Clause 105: The method of clause 101, wherein the step of coding the ground points and the object points further comprises the steps of quantizing the ground points and the object points by a scaling factor, and quantizing the top and bottom thresholds by the scaling factor.

[0270] Clause 106: The method of clause 101, wherein determining the top and bottom thresholds comprises determining top and bottom thresholds for a group of pictures (GOP) including a plurality of frames including the point cloud.

[0271] Clause 107: The method of clause 101, wherein determining the top and bottom thresholds comprises determining top and bottom thresholds for a sequence parameter set (SPS) corresponding to a plurality of frames including the point cloud.

[0272] Clause 108: The method of clause 106, wherein determining the top and bottom thresholds comprises determining top and bottom thresholds for a first frame of the plurality of frames.

[0273] Clause 109: The method of clause 106, wherein determining the top and bottom thresholds comprises determining the top and bottom thresholds as a weighted average of thresholds for a plurality of frames.

[0274] Clause 110: The method of clause 101, further comprising the step of coding a data structure including data representative of at least one of the top threshold or the bottom threshold.

[0275] Clause 111: The method of clause 110, wherein coding the data structure comprises coding at least one of a sequence parameter set (SPS), a geometry parameter set (GPS), or a geometry data unit header (GDH).

[0276] Clause 112: The method of clause 110, wherein coding the data structure comprises coding a value for a top threshold and a flag indicating whether the data should be coded for a bottom threshold.

[0277] Clause 113: The method of clause 110, wherein the step of coding data for at least one of the top threshold and the bottom threshold comprises the step of coding a value of geom_globmotion_threshold0 representing the top threshold and coding a value of geom_globmotion_threshold1 representing the bottom threshold.

[0278] Clause 114: The method of clause 110, wherein coding the data representing at least one of the top threshold or the bottom threshold comprises coding the data representing at least one of the top threshold or the bottom threshold using a respective unsigned integer zeroth order exponential-Golomb value.

[0279] Clause 115: The method of clause 110, wherein coding the data representing at least one of the top threshold or the bottom threshold comprises coding the data representing at least one of the top threshold or the bottom threshold using a respective signed integer zeroth order exponential-Golomb value.

[0280] Clause 116: The method of clause 110, wherein the step of coding data representative of at least one of the top or bottom thresholds comprises the step of coding the data representative of at least one of the top or bottom thresholds using a respective signed fixed length value, and the method further comprises the step of coding data representative of a number of bits allocated to at least one of the top or bottom thresholds.

[0281] Clause 117: The method of clause 110, wherein the step of coding data representative of at least one of the top threshold or the bottom threshold comprises the steps of coding data representative of a midpoint between the top and bottom thresholds, and coding data representative of distances from the midpoint to the top and bottom thresholds.

[0282] Clause 118: The method of clause 110, further comprising the step of coding a geometry data unit header (GDH) including data that overwrites or refines data of the data structure for at least one of the top threshold or the bottom threshold.

[0283] Clause 119: The method of clause 101, wherein the step of determining the top and bottom thresholds comprises the steps of determining a maximum histogram height value, max_box_t, determining a minimum histogram height value, min_box_t, determining a histogram scale value, hist_scale, determining a histogram bin size value, hist_bin_size, according to int((max_box_t - min_box_t) / hist_scale), generating a histogram of points with height values ​​in the range from min_box_t to max_box_t, determining a bin in the histogram which has a maximum number of height values, and determining the top and bottom thresholds according to an offset from the bin which has the maximum number of height values, the offset being defined according to a respective multiple of the standard deviation.

[0284] Clause 120: A device for coding point cloud data, the device comprising: a memory configured to store data representing points of the point cloud; and one or more processors implemented in a circuit and configured to determine elevation values ​​of points in the point cloud, classify the points into a set of ground points or a set of object points according to the elevation values, and code the ground points and the object points according to the classification.

[0285] Clause 121: The device of clause 120, wherein to code the object points, the one or more processors are configured to derive a set of global motion information for the object points and predict the object points using the set of global motion information.

[0286] Clause 122: The device of clause 121, wherein the one or more processors are configured to derive a set of global motion information for only the points of the object.

[0287] Clause 123: The device of clause 121, wherein the set of global motion information comprises a first set of global motion information, and to code the ground points, the one or more processors are configured to derive a second set of global motion information for the ground points and predict the ground points using the second set of global motion information.

[0288] Clause 124: The device of clause 123, wherein the one or more processors are configured to derive a second set of global motion information for only the ground points.

[0289] Clause 125: The device of clause 121, wherein to derive the set of global motion information, the one or more processors are configured to derive a rotation matrix and a translation vector, and to code the object points, the one or more processors are configured to apply the rotation matrix and the translation vector to reference points of a reference frame.

[0290] Clause 126: The device of clause 125, wherein to code the object points, the one or more processors are further configured to determine local node motion vectors for nodes of the prediction tree, the nodes including respective sets of reference points of the reference frame, and to apply the local node motion vectors to the nodes.

[0291] Clause 127: The device of clause 120, wherein to classify the points, the one or more processors are configured to determine a top threshold and a bottom threshold, and to classify points having height values ​​between the top threshold and the bottom threshold into a set of ground points, and to classify points having height values ​​above the top threshold or below the bottom threshold into a set of object points.

[0292] Clause 128: The device of clause 127, wherein the top threshold comprises z_maxi and the bottom threshold comprises z_mini in the range {(x_mini, x_maxi), (y_mini, y_maxi), (z_mini, z_maxi)} of the i-th value.

[0293] Clause 129: The device of clause 128, wherein the i-th value range comprises an i-th value range of N value ranges.

[0294] Clause 130: The device of clause 128, wherein x_mini and y_mini have values ​​of negative infinity and x_maxi and y_maxi have values ​​of infinity.

[0295] Clause 131: The device of clause 127, wherein to classify the ground points and the object points, the one or more processors are further configured to quantize the ground points and the object points by a scaling factor and to quantize the top and bottom thresholds by the scaling factor.

[0296] Clause 132: The device of clause 127, wherein the one or more processors are further configured to code a data structure including data representative of at least one of the top threshold value or the bottom threshold value.

[0297] Clause 133: A method for coding point cloud data, the method comprising the steps of determining elevation values ​​of points in the point cloud, classifying the points into a set of ground points or a set of object points according to the elevation values, and coding the ground points and the object points according to the classification.

[0298] Clause 134: The method of clause 133, wherein coding the object points comprises deriving a set of global motion information for the object points and predicting the object points using the set of global motion information.

[0299] Clause 135: The method of clause 134, wherein deriving a set of global motion information comprises deriving a set of global motion information for only the object points.

[0300] Clause 136: The method of any of clauses 134 and 135, wherein the set of global motion information comprises a first set of global motion information, and coding the ground points comprises deriving a second set of global motion information for the ground points, and predicting the ground points using the second set of global motion information.

[0301] Clause 137: The method of clause 136, wherein deriving a second set of global motion information comprises deriving a second set of global motion information for only the ground points.

[0302] Clause 138: Any of the methods of clauses 134 to 137, wherein the step of deriving a set of global motion information comprises the step of deriving a rotation matrix and a translation vector, and the step of coding the object points comprises the step of applying the rotation matrix and the translation vector to reference points of the reference frame.

[0303] Clause 139: The method of clause 138, wherein the step of coding the object points further comprises the steps of determining local node motion vectors for nodes of the predictive tree, the nodes including respective sets of reference points of a reference frame, and applying the local node motion vectors to the nodes.

[0304] Clause 140: Any of the methods of clauses 133 to 139, wherein the step of classifying the points comprises the steps of determining a top threshold and a bottom threshold, classifying points having height values ​​between the top threshold and the bottom threshold into a set of ground points, and classifying points having height values ​​above the top threshold or below the bottom threshold into a set of object points.

[0305] Clause 141: The method of clause 140, wherein the top threshold comprises z_maxi and the bottom threshold comprises z_mini in the range {(x_mini, x_maxi), (y_mini, y_maxi), (z_mini, z_maxi)} of the i-th value.

[0306] Clause 142: The method of clause 141, wherein the i-th value range comprises an i-th value range of the N value ranges.

[0307] Clause 143: The method of any of clauses 141 and 142, wherein x_mini and y_mini have the value of negative infinity, and x_maxi and y_maxi have the value of infinity.

[0308] Clause 144: The method of any of clauses 134 to 143, wherein the step of coding the ground points and the object points further comprises the steps of quantizing the ground points and the object points by a scaling factor, and quantizing the top and bottom thresholds by the scaling factor.

[0309] Clause 145: The method of any of clauses 140 to 144, wherein determining the top and bottom thresholds comprises determining top and bottom thresholds for a group of pictures (GOP) including a plurality of frames including the point cloud.

[0310] Clause 146: The method of any of clauses 140 to 145, wherein determining the top and bottom thresholds comprises determining top and bottom thresholds for a sequence parameter set (SPS) corresponding to a plurality of frames including the point cloud.

[0311] Clause 147: The method of any of clauses 140 to 144, wherein determining the top and bottom thresholds comprises determining top and bottom thresholds for a first frame of the plurality of frames.

[0312] Clause 148: The method of any of clauses 140 to 146, wherein determining the top and bottom thresholds comprises determining the top and bottom thresholds as a weighted average of thresholds for a plurality of frames.

[0313] Clause 149: The method of any of clauses 140 to 148, further comprising the step of coding a data structure including data representative of at least one of the top threshold or the bottom threshold.

[0314] Clause 150: The method of clause 149, wherein coding the data structure comprises coding at least one of a sequence parameter set (SPS), a geometry parameter set (GPS), or a geometry data unit header (GDH).

[0315] Clause 151: The method of any of clauses 149 and 150, wherein the step of coding the data structure comprises the step of coding a value for a top threshold and a flag indicating whether the data should be coded for a bottom threshold.

[0316] Clause 152: Any of the methods of clauses 149 to 151, wherein the step of coding data for at least one of the top threshold and the bottom threshold comprises the step of coding a value of geom_globmotion_threshold0 representing the top threshold and coding a value of geom_globmotion_threshold1 representing the bottom threshold.

[0317] Clause 153: The method of any of clauses 149 to 152, wherein the step of coding the data representing at least one of the top threshold or the bottom threshold comprises the step of coding the data representing at least one of the top threshold or the bottom threshold using a respective unsigned integer zeroth order exponential Golomb value.

[0318] Clause 154: The method of any of clauses 149 to 152, wherein the step of coding the data representing at least one of the top threshold or the bottom threshold comprises the step of coding the data representing at least one of the top threshold or the bottom threshold using a respective signed integer zeroth order exponential-Golomb value.

[0319] Clause 155: The method of any of clauses 149 to 152, wherein the step of coding data representing at least one of the top or bottom thresholds comprises the step of coding the data representing at least one of the top or bottom thresholds using a respective signed fixed length value, and the method further comprises the step of coding data representing a number of bits allocated to at least one of the top or bottom thresholds.

[0320] Clause 156: The method of any of clauses 149 to 155, wherein the step of coding data representing at least one of the top threshold or the bottom threshold comprises the steps of coding data representing a midpoint between the top and bottom thresholds, and coding data representing distances from the midpoint to the top and bottom thresholds.

[0321] Clause 157: The method of any of clauses 149 to 156, further comprising the step of coding a geometry data unit header (GDH) including data that overwrites or refines data of the data structure for at least one of the top threshold or the bottom threshold.

[0322] Clause 158: Any of the methods of clauses 133 to 157, wherein the step of determining the top and bottom thresholds comprises the steps of determining a maximum histogram height value, max_box_t, determining a minimum histogram height value, min_box_t, determining a histogram scale value, hist_scale, determining a histogram bin size value, hist_bin_size, according to int((max_box_t - min_box_t) / hist_scale), generating a histogram of points with height values ​​in the range from min_box_t to max_box_t, determining the bin in the histogram that has the maximum number of height values, and determining the top and bottom thresholds according to an offset from the bin that has the maximum number of height values, the offset being defined according to a respective multiple of the standard deviation.

[0323] Clause 159: A device for coding point cloud data, comprising: a memory configured to store data representing points of the point cloud; and one or more processors implemented in a circuit and configured to determine elevation values ​​of points in the point cloud, classify the points into a set of ground points or a set of object points according to the elevation values, and code the ground points and the object points according to the classification.

[0324] Clause 160: The device of clause 159, wherein, to code the object points, the one or more processors are configured to derive a set of global motion information for the object points and predict the object points using the set of global motion information.

[0325] Clause 161: The device of clause 160, wherein the one or more processors are configured to derive a set of global motion information for only the points of the object.

[0326] Clause 162: The device of any of clauses 160 and 161, wherein the set of global motion information comprises a first set of global motion information, and to code the ground points, the one or more processors are configured to derive a second set of global motion information for the ground points and predict the ground points using the second set of global motion information.

[0327] Clause 163: The device of clause 162, wherein the one or more processors are configured to derive a second set of global motion information for only the ground points.

[0328] Clause 164: The device of any of clauses 160 to 163, wherein to derive the set of global motion information, the one or more processors are configured to derive a rotation matrix and a translation vector, and to code the object points, the one or more processors are configured to apply the rotation matrix and the translation vector to reference points of the reference frame.

[0329] Clause 165: The device of clause 164, wherein to code the object points, the one or more processors are further configured to determine local node motion vectors for nodes of the prediction tree, the nodes including respective sets of reference points of the reference frame, and to apply the local node motion vectors to the nodes.

[0330] Clause 166: The device of any of clauses 159 to 165, wherein, to classify the points, the one or more processors are configured to determine a top threshold and a bottom threshold, and to classify points having height values ​​between the top threshold and the bottom threshold into a set of ground points, and to classify points having height values ​​above the top threshold or below the bottom threshold into a set of object points.

[0331] Clause 167: The device of clause 166, wherein the top threshold comprises z_maxi and the bottom threshold comprises z_mini in the range {(x_mini, x_maxi), (y_mini, y_maxi), (z_mini, z_maxi)} of the i-th value.

[0332] Clause 168: The device of clause 167, wherein the i-th value range comprises an i-th value range of N value ranges.

[0333] Clause 169: The device of any of clauses 167 and 168, wherein x_mini and y_mini have values ​​of negative infinity and x_maxi and y_maxi have values ​​of infinity.

[0334] Clause 170: The device of any of clauses 166 to 169, wherein, to classify the ground points and the object points, the one or more processors are further configured to quantize the ground points and the object points by a scaling factor and to quantize the top and bottom thresholds by the scaling factor.

[0335] Clause 171: The device of any of clauses 166 to 170, wherein the one or more processors are further configured to code a data structure including data representative of at least one of the top threshold or the bottom threshold.

[0336] It should be appreciated that in some examples, some acts or events of any of the techniques described herein may be performed in a different order, or may be added, combined, or omitted entirely (e.g., not all acts or events described are necessary to the practice of the techniques). Moreover, in some examples, acts or events may be performed simultaneously rather than sequentially, for example, through multi-threaded processing, interrupt processing, or multiple processors.

[0337] In one or more examples, the functions described may be implemented as hardware, software, firmware, or any combination thereof. If implemented in software, the functions may be stored on or transmitted through a computer-readable medium as one or more instructions or code and executed by a hardware-based processing unit. A computer-readable medium may include a computer-readable storage medium corresponding to a tangible medium, such as a data storage medium, or a communication medium including any medium that facilitates transfer of a computer program from one place to another, for example according to a communication protocol. As such, a computer-readable medium may generally correspond to (1) a tangible computer-readable storage medium that is non-transitory, or (2) a communication medium such as a signal or carrier wave. A data storage medium may be any available medium that can be accessed by one or more computers or one or more processors to retrieve instructions, code and / or data structures for implementation of the techniques described in this disclosure. A computer program product may include a computer-readable medium.

[0338] By way of example and not limitation, such computer-readable storage media may comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, flash memory, or any other medium that can be used to store desired program code in the form of instructions or data structures and that can be accessed by a computer. Also, any connection is properly referred to as a computer-readable medium. For example, if instructions are transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of media. However, it should be understood that computer-readable storage media and data storage media do not include connections, carrier waves, signals, or other transitory media, but instead cover non-transitory tangible storage media. As used herein, disk and disc include compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), floppy disk and Blu-ray disc, where disks typically reproduce data magnetically and discs reproduce data optically using lasers. Combinations of the above should also be included within the scope of computer readable media.

[0339] The instructions may be executed by one or more processors, such as one or more digital signal processors (DSPs), general-purpose microprocessors, application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), or other equivalent integrated or discrete logic circuitry. Thus, the terms "processor" and "processing circuitry" as used herein may refer to any of the above structures, or any other structures suitable for implementing the techniques described herein. In addition, in some aspects, the functions described herein may be provided within dedicated hardware and / or software modules configured for encoding and decoding, or may be incorporated into a composite codec. Also, the techniques may be fully implemented in one or more circuits or logic elements.

[0340] The techniques of the present disclosure may be implemented in a wide variety of devices or apparatuses, including wireless handsets, integrated circuits (ICs), or sets of ICs (e.g., chipsets). Although various components, modules, or units have been described in this disclosure to highlight functional aspects of devices configured to perform the disclosed techniques, they do not necessarily require realization by different hardware units. Rather, as described above, the various units may be combined in a codec hardware unit or may be provided by a collection of interoperable hardware units, including one or more processors as described above, along with appropriate software and / or firmware.

[0341] Various examples have been described. These and other examples are within the scope of the following claims. [Explanation of symbols]

[0342] 102 Source Device 104 Data Sources 106 Memory 108 Output Interface 110 Computer-readable medium 112 Storage Devices 114 File Server 116 Destination Device 118 Data Consumer 120 Memory 122 Input Interface 130 Reference group 132 Global Motion Vectors 134 predicted groups 136 Reference Nodes 140 Current group 142 Current Node 146 bitstream 150 reference group 152 Global Motion Vectors 154 predicted groups 156 Reference Nodes 160 Current group 162 Current Node 164 Vector 166 bit stream 200G-PCC Encoder 202 Coordinate Transformation Unit 204 Color Conversion Unit 206 Voxelization Unit 208 Attribute Transfer Unit 210 8-tree analysis unit 212 Surface Approximation Analysis Unit 214 Arithmetic Coding Unit 216 Geometry Reconstruction Unit 218 RAHT Unit 220 LOD generation units 222 Lifting Unit 224 Coefficient Quantization Unit 226 Arithmetic Coding Unit 228 Memory 300 G-PCC Decoder 302 Geometry Arithmetic Decoding Unit 304 Attribute Arithmetic Decoding Unit 306 8th Tree Synthesis Unit 308 Inverse Quantization Unit 310 Surface Approximation Synthesis Unit 312 Geometry Reconstruction Unit 314 RAHT Unit 316 LOD generation units 318 Reverse Lifting Unit 320 Inverse Coordinate Transformation Unit 322 Inverse Color Conversion Unit 324 Memory 350 Prediction Trees 352 nodes 354 nodes 356 nodes 358 nodes 360 nodes 380 LIDAR 382 Laser 384 Radius 386 Azimuth 388 Laser Index 452 z_top 454 z_bottom 462 z_bottom 464 z_top 470 point cloud 472 Object Points 474 Road Points 476 points 600 Laser Package 602 Head 604 Laser 606 Laser Beam 608 vertical axis 610 Arrow 612 Arrow 902 Lighting equipment 904 Sensors 906 light 908 Object 910 light 911 Lens 912 images 918 signal 920 point cloud generator 1000 vehicles 1002 Laser Package 1004 Laser Beam 1006 Pedestrians 1008 Bitstream 1010 Vehicle 1012 Server System 1100 First User 1102 1st position 1104 XR Headset 1106 Object 1108 Bitstream 1110 XR Headset 1112 users 1114 2nd position 1200 Mobile Devices 1202 Object 1204 Bitstream 1206 Remote Device

Claims

1. A method for encoding point cloud data, comprising: determining elevation values ​​of points in the point cloud; classifying the points into a set of ground points or a set of object points according to the height values; encoding the ground points and the object points according to the classification; Equipped with wherein encoding the object points comprises deriving a first set of global motion information for the object points; and predicting the object points using the first set of global motion information; wherein encoding the ground points comprises deriving a second set of global motion information for the ground points; and predicting the ground points using the second set of global motion information. method.

2. 1. A method for decoding point cloud data, comprising: determining elevation values ​​of points in the point cloud; classifying the points into a set of ground points or a set of object points according to the height values; decoding the ground points and the object points according to the classification; Equipped with wherein encoding the object points comprises deriving a first set of global motion information for the object points; and predicting the object points using the first set of global motion information; wherein encoding the ground points comprises deriving a second set of global motion information for the ground points; and predicting the ground points using the second set of global motion information. method.

3. The step of classifying the points further comprises: determining a top threshold and a bottom threshold; classifying points having height values ​​between the top and bottom thresholds into the set of ground points; classifying points having height values ​​above the top threshold or below the bottom threshold into a set of body points; The method of claim 1 , comprising:

4. The method of claim 3 , wherein determining the top and bottom thresholds comprises determining the top and bottom thresholds for a group of pictures (GOP) that includes a plurality of frames that include the point cloud.

5. a data structure including data representative of the top and bottom thresholds; decoding data representative of a midpoint between said top and bottom thresholds; decoding data from said midpoint representing the distance between said top threshold and said bottom threshold; The method of claim 3, further comprising the step of decrypting by

6. A device for encoding point cloud data, comprising: a memory configured to store data representing points of the point cloud; One or more processors implemented in circuitry, determining elevation values ​​of points in the point cloud; classifying said points into a set of ground points or a set of object points according to said height values; coding the ground points and the object points according to the classification; one or more processors, Equipped with encoding the object points comprises deriving a first set of global motion information for the object points; and predicting the object points using the first set of global motion information; encoding the ground points comprises deriving a second set of global motion information for the ground points; and predicting the ground points using the second set of global motion information. device.

7. 1. A device for decoding point cloud data, comprising: a memory configured to store data representing points of the point cloud; One or more processors implemented in circuitry, determining elevation values ​​of points in the point cloud; classifying said points into a set of ground points or a set of object points according to said height values; coding the ground points and the object points according to the classification; one or more processors, Equipped with encoding the object points comprises deriving a first set of global motion information for the object points; and predicting the object points using the first set of global motion information; A device, wherein encoding the ground points comprises deriving a second set of global motion information for the ground points and predicting the ground points using the second set of global motion information.

8. To derive the set of global motion information, the one or more processors are configured to derive a rotation matrix and a translation vector, and to code the object points, the one or more processors are configured to: applying the rotation matrix and the translation vector to a reference point of a reference frame; determining local node motion vectors for nodes of a prediction tree, the nodes including a respective set of reference points of the reference frames; applying the local node motion vector to the node; The device of claim 7 , configured to:

9. To classify the points, the one or more processors: determining a top threshold and a bottom threshold; classifying points having height values ​​between said top threshold and said bottom threshold into said set of ground points; Classifying points having height values ​​above the top threshold or below the bottom threshold into a set of body points. The device of claim 6 , configured to:

10. The top threshold is the range of the i-th value {(x_min i , x_max i ), (y_min i , y_max i ), (z_min i , z_max i )}'s z_max i and the bottom threshold is z_min i 10. The device of claim 9, wherein the i-th value range comprises an i-th value range of N value ranges.

11. To classify the ground points and the object points, the one or more processors further comprise: quantizing the ground points and the object points by a scaling factor; quantizing the top and bottom thresholds by the scaling factor The device of claim 9 , configured to:

12. 10. The device of claim 9, wherein the one or more processors are further configured to decode a data structure including data representing at least one of the top threshold value or the bottom threshold value.

13. The method of claim 12, further comprising: decoding the data representing at least one of the top threshold value or the bottom threshold value. decoding data representative of a midpoint between the top and bottom thresholds; decoding data from said midpoint representing a distance between said top threshold value and said bottom threshold value; The device of claim 12, comprising:

14. The device described in claim 12, wherein the one or more processors are further configured to decode a geometry data unit header (GDH) including data that overwrites or refines the data of the data structure for at least one of the top threshold or the bottom threshold.

15. A computer readable storage medium having instructions stored thereon which, when executed, cause a processor to perform the method of any one of claims 1 to 5.