Method, apparatus, and medium for video processing
By pre-processing and post-processing 3D Gaussian Splatting models to convert floating-point data to integers and applying G-PCC coding tools, the solution addresses the inefficiencies in existing G-PCC for 3DGS attributes, achieving enhanced compression efficiency.
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- DOUYIN VISION CO LTD
- Filing Date
- 2025-10-31
- Publication Date
- 2026-05-07
AI Technical Summary
Existing point cloud coding techniques, particularly Geometry-based Point Cloud Compression (G-PCC), lack efficient support for 3D Gaussian Splatting (3DGS) model attributes such as scale, rotation, and spherical harmonic coefficients, and require pre-processing of floating-point data to integer precision for compression.
Implement pre-processing to quantize floating-point values of geometry and attribute information to integers before encoding, and apply G-PCC coding tools, followed by post-processing to restore floating-point precision after decoding, utilizing methods like Octree-based Geometry Coding, RAHT, and Lifting Transform for efficient compression of 3DGS models.
Enhances the compression efficiency of 3D Gaussian Splatting models by supporting specialized attributes and handling floating-point data effectively, resulting in improved rate-distortion performance.
Smart Images

Figure CN2025131914_07052026_PF_FP_ABST
Abstract
Description
METHOD, APPARATUS, AND MEDIUM FOR VIDEO PROCESSINGFIELDS
[0001] Embodiments of the present disclosure relate generally to point cloud coding techniques, and more particularly, to Gaussian splatting (GS) model based video processing using geometry-based point cloud compression (G-PCC) .BACKGROUND
[0002] A point cloud is a collection of individual data points in a three-dimensional (3D) plane with each point having a set coordinate on the X, Y, and Z axes. Thus, a point cloud may be used to represent the physical content of the three-dimensional space. Point clouds have shown to be a promising way to represent 3D visual data for a wide range of immersive applications, from augmented reality to autonomous cars.
[0003] Point cloud coding standards have evolved primarily through the development of the well-known MPEG organization. MPEG, short for Moving Picture Experts Group, is one of the main standardization groups dealing with multimedia. In 2017, the MPEG 3D Graphics Coding group (3DG) published a call for proposals (CFP) document to start to develop point cloud coding standard. The final standard will consist in two classes of solutions. Video-based Point Cloud Compression (V-PCC or VPCC) is appropriate for point sets with a relatively uniform distribution of points. Geometry-based Point Cloud Compression (G-PCC or GPCC) is appropriate for more sparse distributions. However, coding efficiency of conventional point cloud coding techniques is generally expected to be further improved.SUMMARY
[0004] Embodiments of the present disclosure provide a solution for video processing.
[0005] In a first aspect, a method for video processing is proposed. The method comprises: obtaining, for a conversion between a current Gaussian splatting (GS) model of a video and a bitstream of the video, geometry information and attribute information of the current GS model, the current GS model being represented by positions in a space with the attribute information associated with the positions, the geometry information comprising the positions; pre-processing the geometry information and the attribute information to quantize floating-point values of the geometry information and the attribute information to integer values; and performing the conversion by applying at least one coding tool for geometry-based point cloud compression (G-PCC) to the pre-processed geometry information and attribute information. The method in accordance with the first aspect of the present disclosure pre-prosses the floating-point precision data for both geometry and attribute of the GS model before coding to accommodate G-PCC’s compression algorithms.
[0006] In a second aspect, another method for video processing is proposed. The method comprises: obtaining, for a conversion between a current Gaussian splatting (GS) model of a video and a bitstream of the video, geometry information and attribute information of the current GS model, the current GS model being represented by positions in a space with the attribute information associated with the positions, the geometry information comprising the positions; applying at least one coding tool for geometry-based point cloud compression (G-PCC) to at least one of the geometry information or the attribute information; and performing the conversion based on the applying, where the attribute information comprises at least one of: scale information, rotation information, spherical harmonic (SH) coefficients or opacity information of the current GS model. The method in accordance with the second aspect of the present disclosure supports different types of attribute information of GS model, such as SH coefficients, scale, rotation and opacity.
[0007] In a third aspect, an apparatus for video processing is proposed. The apparatus comprises a processor and a non-transitory memory with instructions thereon. The instructions upon execution by the processor, cause the processor to perform a method in accordance with the first, or second aspect of the present disclosure.
[0008] In a fourth aspect, a non-transitory computer-readable storage medium is proposed. The non-transitory computer-readable storage medium stores instructions that cause a processor to perform a method in accordance with the first, or second aspect of the present disclosure.
[0009] In a fifth aspect, another non-transitory computer-readable recording medium is proposed. The non-transitory computer-readable recording medium stores a bitstream of a video which is generated by a method performed by a video processing apparatus. The method comprises: obtaining geometry information and attribute information of a current Gaussian splatting (GS) model of the video, the current GS model being represented by positions in a space with the attribute information associated with the positions, the geometry information comprising the positions; pre-processing the geometry information and the attribute information to quantize floating-point values of the geometry information and the attribute information to integer values; and generating the bitstream by applying at least one coding tool for geometry-based point cloud compression (G-PCC) to the pre-processed geometry information and attribute information.
[0010] In a sixth aspect, another non-transitory computer-readable recording medium is proposed. The non-transitory computer-readable recording medium stores a bitstream of a video which is generated by a method performed by a video processing apparatus. The method comprises: obtaining geometry information and attribute information of a current Gaussian splatting (GS) model of the video, the current GS model being represented by positions in a space with the attribute information associated with the positions, the geometry information comprising the positions, wherein the attribute information comprises at least one of: scale information, rotation information, spherical harmonic (SH) coefficients or opacity information of the current GS model; applying at least one coding tool for geometry-based point cloud compression (G-PCC) to at least one of the geometry information or the attribute information; and generating the bitstream based on the applying.
[0011] In a seventh aspect, a method for storing a bitstream of a video is proposed. The method comprises: obtaining geometry information and attribute information of a current Gaussian splatting (GS) model of the video, the current GS model being represented by positions in a space with the attribute information associated with the positions, the geometry information comprising the positions; pre-processing the geometry information and the attribute information to quantize floating-point values of the geometry information and the attribute information to integer values; generating the bitstream by applying at least one coding tool for geometry-based point cloud compression (G-PCC) to the pre-processed geometry information and attribute information; and storing the bitstream in a non-transitory computer-readable recording medium.
[0012] In an eighth aspect, a method for storing a bitstream of a video is proposed. The method comprises: obtaining geometry information and attribute information of a current Gaussian splatting (GS) model of the video, the current GS model being represented by positions in a space with the attribute information associated with the positions, the geometry information comprising the positions, wherein the attribute information comprises at least one of: scale information, rotation information, spherical harmonic (SH) coefficients or opacity information of the current GS model; applying at least one coding tool for geometry-based point cloud compression (G-PCC) to at least one of the geometry information or the attribute information; generating the bitstream based on the applying; and storing the bitstream in a non-transitory computer-readable recording medium.
[0013] This Summary is provided to introduce a selection of concepts in a simplified form that are further described below in the Detailed Description. This Summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.BRIEF DESCRIPTION OF THE DRAWINGS
[0014] Through the following detailed description with reference to the accompanying drawings, the above and other objectives, features, and advantages of example embodiments of the present disclosure will become more apparent. In the example embodiments of the present disclosure, the same reference numerals usually refer to the same components.
[0015] Fig. 1 illustrates a block diagram of an example point cloud coding system in accordance with some embodiments of the present disclosure;
[0016] Fig. 2 illustrates a block diagram of an example GPCC encoder in accordance with some embodiments of the present disclosure;
[0017] Fig. 3 illustrates a block diagram of an example GPCC decoder in accordance with some embodiments of the present disclosure;
[0018] Fig. 4 illustrates an example of Gaussian coding pipeline using video codec;
[0019] Fig. 5 illustrates an example of Gaussian coding pipeline using dedicated codec;
[0020] Fig. 6 illustrates an example architecture showing main modules highlighted so far in the approaches discussed for Gaussian coding;
[0021] Fig. 7 illustrates an example of compressing Gaussian Splatting model using geometry-based point cloud compression in accordance with embodiments of the present disclosure;
[0022] Fig. 8 illustrates a flowchart of a method for video processing in accordance with embodiments of the present disclosure;
[0023] Fig. 9 illustrates a flowchart of another method for video processing in accordance with embodiments of the present disclosure; and
[0024] Fig. 10 illustrates a block diagram of a computing device in which various embodiments of the present disclosure can be implemented.
[0025] Throughout the drawings, the same or similar reference numerals usually refer to the same or similar elements.DETAILED DESCRIPTION
[0026] Principle of the present disclosure will now be described with reference to some embodiments. It is to be understood that these embodiments are described only for the purpose of illustration and help those skilled in the art to understand and implement the present disclosure, without suggesting any limitation as to the scope of the disclosure. The disclosure described herein can be implemented in various manners other than the ones described below.
[0027] In the following description and claims, unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skills in the art to which this disclosure belongs.
[0028] References in the present disclosure to “one embodiment, ” “an embodiment, ” “an example embodiment, ” and the like indicate that the embodiment described may include a particular feature, structure, or characteristic, but it is not necessary that every embodiment includes the particular feature, structure, or characteristic. Moreover, such phrases are not necessarily referring to the same embodiment. Further, when a particular feature, structure, or characteristic is described in connection with an example embodiment, it is submitted that it is within the knowledge of one skilled in the art to affect such feature, structure, or characteristic in connection with other embodiments whether or not explicitly described.
[0029] It shall be understood that although the terms “first” and “second” etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, a first element could be termed a second element, and similarly, a second element could be termed a first element, without departing from the scope of example embodiments. As used herein, the term “and / or” includes any and all combinations of one or more of the listed terms.
[0030] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments. As used herein, the singular forms “a” , “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms “comprises” , “comprising” , “has” , “having” , “includes” and / or “including” , when used herein, specify the presence of stated features, elements, and / or components etc., but do not preclude the presence or addition of one or more other features, elements, components and / or combinations thereof. Example Environment
[0031] Fig. 1 is a block diagram that illustrates an example point cloud coding system 100 that may utilize the techniques of the present disclosure. As shown, the point cloud coding system 100 may include a source device 110 and a destination device 120. The source device 110 can be also referred to as a point cloud encoding device, and the destination device 120 can be also referred to as a point cloud decoding device. In operation, the source device 110 can be configured to generate encoded point cloud data and the destination device 120 can be configured to decode the encoded point cloud data generated by the source device 110. The techniques of some example embodiments of this disclosure are generally directed to coding (encoding and / or decoding) point cloud data, i.e., to support point cloud compression. The coding may be effective in compressing and / or decompressing point cloud data.
[0032] Source device 110 and destination device 120 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 and mobile phones, televisions, cameras, display devices, digital media players, video gaming consoles, video streaming devices, vehicles (e.g., terrestrial or marine vehicles, spacecraft, aircraft, etc. ) , robots, LIDAR devices, satellites, extended reality devices, or the like. In some cases, source device 110 and destination device 120 may be equipped for wireless communication.
[0033] The source device 110 may include a data source 112, a memory 114, a GPCC encoder 116, and an input / output (I / O) interface 118. The destination device 120 may include an input / output (I / O) interface 128, a GPCC decoder 126, a memory 124, and a data consumer 122. In accordance with some example embodiments of this disclosure, GPCC encoder 116 of source device 110 and GPCC decoder 126 of destination device 120 may be configured to apply the techniques of some example embodiments of this disclosure related to point cloud coding. Thus, source device 110 represents an example of an encoding device, while destination device 120 represents an example of a decoding device. In other examples, source device 110 and destination device 120 may include other components or arrangements. For example, source device 110 may receive data (e.g., point cloud data) from an internal or external source. Likewise, destination device 120 may interface with an external data consumer, rather than include a data consumer in the same device.
[0034] In general, data source 112 represents a source of point cloud data (i.e., raw, unencoded point cloud data) and may provide a sequential series of “frames” of the point cloud data to GPCC encoder 116, which encodes point cloud data for the frames. In some examples, data source 112 generates the point cloud data. Data source 112 of source device 110 may include a point cloud capture device, such as any of a variety of cameras or sensors, e.g., one or more video cameras, an archive containing previously captured point cloud data, a 3D scanner or a light detection and ranging (LIDAR) device, and / or a data feed interface to receive point cloud data from a data content provider. Thus, in some examples, data source 112 may generate the point cloud data based on signals from a LIDAR apparatus. Alternatively or additionally, point cloud data may be computer-generated from scanner, camera, sensor or other data. For example, data source 112 may generate the point cloud data, or produce a combination of live point cloud data, archived point cloud data, and computer-generated point cloud data. In each case, GPCC encoder 116 encodes the captured, pre-captured, or computer-generated point cloud data. GPCC encoder 116 may rearrange frames of the point cloud data from the received order (sometimes referred to as “display order” ) into a coding order for coding. GPCC encoder 116 may generate one or more bitstreams including encoded point cloud data. Source device 110 may then output the encoded point cloud data via I / O interface 118 for reception and / or retrieval by, e.g., I / O interface 128 of destination device 120. The encoded point cloud data may be transmitted directly to destination device 120 via the I / O interface 118 through the network 130A. The encoded point cloud data may also be stored onto a storage medium / server 130B for access by destination device 120.
[0035] Memory 114 of source device 110 and memory 124 of destination device 120 may represent general purpose memories. In some examples, memory 114 and memory 124 may store raw point cloud data, e.g., raw point cloud data from data source 112 and raw, decoded point cloud data from GPCC decoder 126. Additionally or alternatively, memory 114 and memory 124 may store software instructions executable by, e.g., GPCC encoder 116 and GPCC decoder 126, respectively. Although memory 114 and memory 124 are shown separately from GPCC encoder 116 and GPCC decoder 126 in this example, it should be understood that GPCC encoder 116 and GPCC decoder 126 may also include internal memories for functionally similar or equivalent purposes. Furthermore, memory 114 and memory 124 may store encoded point cloud data, e.g., output from GPCC encoder 116 and input to GPCC decoder 126. In some examples, portions of memory 114 and memory 124 may be allocated as one or more buffers, e.g., to store raw, decoded, and / or encoded point cloud data. For instance, memory 114 and memory 124 may store point cloud data.
[0036] I / O interface 118 and I / O interface 128 may represent wireless transmitters / receivers, modems, wired networking components (e.g., Ethernet cards) , wireless communication components that operate according to any of a variety of IEEE 802.11 standards, or other physical components. In examples where I / O interface 118 and I / O interface 128 comprise wireless components, I / O interface 118 and I / O interface 128 may be configured to transfer data, such as encoded point cloud data, according to a cellular communication standard, such as 4G, 4G-LTE (Long-Term Evolution) , LTE Advanced, 5G, or the like. In some examples where I / O interface 118 comprises a wireless transmitter, I / O interface 118 and I / O interface 128 may be configured to transfer data, such as encoded point cloud data, according to other wireless standards, such as an IEEE 802.11 specification. In some examples, source device 110 and / or destination device 120 may include respective system-on-a-chip (SoC) devices. For example, source device 110 may include an SoC device to perform the functionality attributed to GPCC encoder 116 and / or I / O interface 118, and destination device 120 may include an SoC device to perform the functionality attributed to GPCC decoder 126 and / or I / O interface 128.
[0037] The techniques of some example embodiments 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.
[0038] I / O interface 128 of destination device 120 receives an encoded bitstream from source device 110. The encoded bitstream may include signaling information defined by GPCC encoder 116, which is also used by GPCC decoder 126, such as syntax elements having values that represent a point cloud. Data consumer 122 uses the decoded data. For example, data consumer 122 may use the decoded point cloud data to determine the locations of physical objects. In some examples, data consumer 122 may comprise a display to present imagery based on the point cloud data.
[0039] GPCC encoder 116 and GPCC decoder 126 each may be implemented as any of a variety of suitable encoder and / or decoder circuitry, 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 combinations thereof. When the techniques are implemented partially in software, a device may store instructions for the software in a suitable, non-transitory computer-readable medium and execute the instructions in hardware using one or more processors to perform the techniques of some example embodiments of this disclosure. Each of GPCC encoder 116 and GPCC decoder 126 may be included in one or more encoders or decoders, either of which may be integrated as part of a combined encoder / decoder (CODEC) in a respective device. A device including GPCC encoder 116 and / or GPCC decoder 126 may comprise one or more integrated circuits, microprocessors, and / or other types of devices.
[0040] GPCC encoder 116 and GPCC decoder 126 may operate according to a coding standard, such as video point cloud compression (VPCC) standard or a geometry point cloud compression (GPCC) standard. This disclosure may generally refer to coding (e.g., encoding and decoding) of frames to include the process of encoding or decoding data. An encoded bitstream generally includes a series of values for syntax elements representative of coding decisions (e.g., coding modes) .
[0041] A point cloud may contain a set of points in a 3D space, and may have attributes associated with the point. The attributes may be color information such as R, G, B or Y, Cb, Cr, or reflectance information, or other attributes. Point clouds may be captured by a variety of cameras or sensors such as LIDAR sensors and 3D scanners and may also be computer-generated. Point cloud data are used in a variety of applications including, but not limited to, construction (modeling) , graphics (3D models for visualizing and animation) , and the automotive industry (LIDAR sensors used to help in navigation) .
[0042] Fig. 2 is a block diagram illustrating an example of a GPCC encoder 200, which may be an example of the GPCC encoder 116 in the system 100 illustrated in Fig. 1, in accordance with some embodiments of the present disclosure. Fig. 3 is a block diagram illustrating an example of a GPCC decoder 300, which may be an example of the GPCC decoder 126 in the system 100 illustrated in Fig. 1, in accordance with some embodiments of the present disclosure.
[0043] In both GPCC encoder 200 and GPCC decoder 300, point cloud positions are coded first. Attribute coding depends on the decoded geometry. In Fig. 2 and Fig. 3, the region adaptive hierarchical transform (RAHT) unit 218, surface approximation analysis unit 212, RAHT unit 314 and surface approximation synthesis unit 310 are options typically used for Category 1 data. The level-of-detail (LOD) generation unit 220, lifting unit 222, LOD generation unit 316 and inverse lifting unit 318 are options typically used for Category 3 data. All the other units are common between Categories 1 and 3.
[0044] For Category 3 data, the compressed geometry is typically represented as an octree from the root all the way down to a leaf level of individual voxels. For Category 1 data, the compressed geometry is typically represented by a pruned octree (i.e., an octree from the root down to a leaf level of blocks larger than voxels) plus a model that approximates the surface within each leaf of the pruned octree. In this way, both Category 1 and 3 data share the octree coding mechanism, while Category 1 data may in addition approximate the voxels within each leaf with a surface model. The surface model used is a triangulation comprising 1-10 triangles per block, resulting in a triangle soup. The Category 1 geometry codec is therefore known as the Trisoup geometry codec, while the Category 3 geometry codec is known as the Octree geometry codec.
[0045] In the example of Fig. 2, GPCC encoder 200 may include a coordinate transform unit 202, a color transform unit 204, a voxelization unit 206, an attribute transfer unit 208, an octree analysis unit 210, a surface approximation analysis unit 212, an arithmetic encoding unit 214, a geometry reconstruction unit 216, an RAHT unit 218, a LOD generation unit 220, a lifting unit 222, a coefficient quantization unit 224, and an arithmetic encoding unit 226.
[0046] As shown in the example of Fig. 2, GPCC encoder 200 may receive a set of positions and a set of attributes. The positions may include coordinates of points in a point cloud. The attributes may include information about points in the point cloud, such as colors associated with points in the point cloud.
[0047] Coordinate transform unit 202 may apply a transform to the coordinates of the points to transform the coordinates from an initial domain to a transform domain. This disclosure may refer to the transformed coordinates as transform coordinates. Color transform unit 204 may apply a transform to convert color information of the attributes to a different domain. For example, color transform unit 204 may convert color information from an RGB color space to a YCbCr color space.
[0048] Furthermore, in the example of Fig. 2, voxelization unit 206 may voxelize the transform coordinates. Voxelization of the transform coordinates may include quantizing and removing some points of the point cloud. In other words, multiple points of the point cloud may be subsumed within a single “voxel, ” which may thereafter be treated in some respects as one point. Furthermore, octree analysis unit 210 may generate an octree based on the voxelized transform coordinates. Additionally, in the example of Fig. 2, surface approximation analysis unit 212 may analyze the points to potentially determine a surface representation of sets of the points. Arithmetic encoding unit 214 may perform arithmetic encoding on syntax elements representing the information of the octree and / or surfaces determined by surface approximation analysis unit 212. GPCC encoder 200 may output these syntax elements in a geometry bitstream.
[0049] Geometry reconstruction unit 216 may reconstruct transform coordinates of points in the point cloud based on the octree, data indicating the surfaces determined by surface approximation analysis unit 212, and / or other information. The number of transform coordinates reconstructed by geometry reconstruction unit 216 may be different from the original number of points of the point cloud because of voxelization and surface approximation. This disclosure may refer to the resulting points as reconstructed points. Attribute transfer unit 208 may transfer attributes of the original points of the point cloud to reconstructed points of the point cloud data.
[0050] Furthermore, RAHT unit 218 may apply RAHT coding to the attributes of the reconstructed points. Alternatively or additionally, LOD generation unit 220 and lifting unit 222 may apply LOD processing and lifting, respectively, to the attributes of the reconstructed points. RAHT unit 218 and lifting unit 222 may generate coefficients based on the attributes. Coefficient quantization unit 224 may quantize the coefficients generated by RAHT unit 218 or lifting unit 222. Arithmetic encoding unit 226 may apply arithmetic coding to syntax elements representing the quantized coefficients. GPCC encoder 200 may output these syntax elements in an attribute bitstream.
[0051] In the example of Fig. 3, GPCC decoder 300 may include a geometry arithmetic decoding unit 302, 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, a coordinate inverse transform unit 320, and a color inverse transform unit 322.
[0052] GPCC decoder 300 may obtain a geometry bitstream and an attribute bitstream. Geometry arithmetic decoding unit 302 of decoder 300 may apply arithmetic decoding (e.g., CABAC or other type of arithmetic decoding) to syntax elements in the geometry bitstream. Similarly, attribute arithmetic decoding unit 304 may apply arithmetic decoding to syntax elements in attribute bitstream.
[0053] Octree synthesis unit 306 may synthesize an octree based on syntax elements parsed from geometry bitstream. In instances where surface approximation is used in geometry bitstream, surface approximation synthesis unit 310 may determine a surface model based on syntax elements parsed from geometry bitstream and based on the octree.
[0054] Furthermore, geometry reconstruction unit 312 may perform a reconstruction to determine coordinates of points in a point cloud. Coordinate inverse transform unit 320 may apply an inverse transform to the reconstructed coordinates to convert the reconstructed coordinates (positions) of the points in the point cloud from a transform domain back into an initial domain.
[0055] Additionally, in the example of Fig. 3, inverse quantization unit 308 may inverse quantize attribute values. The attribute values may be based on syntax elements obtained from attribute bitstream (e.g., including syntax elements decoded by attribute arithmetic decoding unit 304) .
[0056] Depending on how the attribute values are encoded, RAHT unit 314 may perform RAHT coding to determine, based on the inverse quantized attribute values, color values for points of the point cloud. Alternatively, LOD generation unit 316 and inverse lifting unit 318 may determine color values for points of the point cloud using a level of detail-based technique.
[0057] Furthermore, in the example of Fig. 3, color inverse transform unit 322 may apply an inverse color transform to the color values. The inverse color transform may be an inverse of a color transform applied by color transform unit 204 of encoder 200. For example, color transform unit 204 may transform color information from an RGB color space to a YCbCr color space. Accordingly, color inverse transform unit 322 may transform color information from the YCbCr color space to the RGB color space.
[0058] The various units of Fig. 2 and Fig. 3 are illustrated to assist with understanding the operations performed by encoder 200 and 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 particular functionality and are preset on the operations that can be performed. Programmable circuits refer to circuits that can be programmed to perform various tasks and provide flexible functionality in the operations that can be performed. For instance, programmable circuits may execute software or firmware that cause the programmable circuits to operate in the manner defined by instructions of the software or firmware. Fixed-function circuits may execute software instructions (e.g., to receive parameters or output parameters) , but the types of operations that the fixed-function circuits perform are generally immutable. In some examples, one or more of the units may be distinct circuit blocks (fixed-function or programmable) , and in some examples, one or more of the units may be integrated circuits.
[0059] Some example embodiments of the present disclosure will be described in detailed hereinafter. It should be understood that section headings are used in the present document to facilitate ease of understanding and do not limit the embodiments disclosed in a section to only that section. Furthermore, while certain embodiments are described with reference to GPCC or other specific point cloud codecs, the disclosed techniques are applicable to other point cloud coding technologies also. Furthermore, while some embodiments describe point cloud coding steps in detail, it will be understood that corresponding steps decoding that undo the coding will be implemented by a decoder. 1.Summary
[0060] This disclosure is related to 3D scene compression and transmission technologies. Specifically, it is related to compress Gaussian Splatting model using geometry-based point cloud compression (G-PCC) . The ideas may be applied individually or in various combination, to any gaussian splatting model coding standard, non-standard gaussian splatting model codec, point cloud coding standard or non-standard point cloud codec, e.g., the being-developed Geometry-based Point Cloud Compression (G-PCC) . 2. Abbreviations
[0061] G-PCC Geometry based Point Cloud Compression
[0062] V-PCC Video-based Point Cloud Compression
[0063] MPEG Moving Picture Experts Group
[0064] WG Working Group
[0065] 3DG 3D Graphics Coding Group
[0066] 3DGS 3D Gaussian Splatting
[0067] 4DGS 4D Gaussian Splatting. 2. Background
[0068] Point cloud coding standards have evolved primarily through the development of the well-known MPEG organization. MPEG, short for Moving Picture Experts Group, is one of the main standardization groups dealing with multimedia. In 2017, the MPEG 3D Graphics Coding group (3DG) published a call for proposals (CFP) document to start to develop point cloud coding standard. The final standard will consist in two classes of solutions. Video-based Point Cloud Compression (V-PCC) is appropriate for point sets with a relatively uniform distribution of points. Geometry-based Point Cloud Compression (G-PCC) is appropriate for more sparse distributions. Both V-PCC and G-PCC support the coding and decoding for single point cloud and point cloud sequence.
[0069] In one point cloud, there may be geometry information and attribute information. Geometry information is used to describe the geometry locations of the data points. Attribute information is used to record some details of the data points, such as textures, normal vectors, reflections and so on. Point cloud codec can process the various information in different ways. Usually there are many optional tools in the codec to support the coding and decoding of geometry information and attribute information respectively.
[0070] In 2024, the Working Group (WG) 7 on MPEG 3D Graphics Coding group (3DG) and WG 4 on MPEG Video Coding both started exploring 3D / 4D gaussian splatting (3DGS / 4DGS) coding techniques. There is one Joint Exploration Experiment 6 (JEE 6) to study the 3D / 4D gaussian splatting compression. The objective of JEE 6 between WG 4 and WG 7 is to conduct a comprehensive investigation into various aspects of 3D / 4D Gaussian representation and compression, considering the video and point cloud perspective.
[0071] 3DGS, as introduced in the paper “3D Gaussian Splatting for Real-Time Radiance Field Rendering” , is a technique used in computer graphics for efficiently representing and rendering 3D scenes. This method allows for the interpolation and reconstruction of scene information across viewpoints, enabling the rendering of novel views in real-time.
[0072] The representation and compression of 3DGS-related data can be viewed at least from two perspectives. One is the video perspective, where 3DGS data is considered as an intermediate representation for encoding and decoding multiview video for 6DoF rendering, i.e., input and output are both videos. The other is the point cloud perspective, where input and output are 3DGS data (3DGS model) represented by positions in 3D space, with associated attributes, like point clouds. 3.1 Video Perspective
[0073] 3DGS for novel view synthesis is considered an emerging technology for immersive video applications. 3DGS has been explored in WG4 since MPEG 145. From the WG4 perspective, 3DGS data is a representation of video for 6DoF rendering. Given multiview or monocular videos as input, it is converted into 3D / 4D Gaussians, and then rendered into output videos at user-specified viewpoints.
[0074] From the video perspective, coding of 3DGS / 4DGS data may include the design of both, representation and the compression. Converting video into 3DGS / 4DGS data is also a part of the encoding pipeline, and rendering the reconstructed 3DGS / 4DGS data into novel view videos is a part of the decoding process. Therefore, the representation and compression may be considered jointly.
[0075] Regarding the representation, WG4 has investigated different variants of 3DGS for representing static scenes, including the original 3DGS representation and other variants. Furthermore, there is a lack of de facto representation for dynamic scenes, making it important to investigate 4D Gaussian representations oriented for the use cases and requirements of immersive video transmission. Finally, it may be valuable to jointly consider representation and compression to achieve the best rate-distortion performance.
[0076] In terms of the compression techniques, WG4 experts have explored various techniques, including using 2D legacy video codec and dedicated codec for 3DGS / 4DGS data. Fig. 4 illustrates an example of Gaussian coding pipeline using video codec. Fig. 5 illustrates an example og Gaussian pipeline using dedicated codec. In the former case, the data may be sorted to fit into 2D image space, and then compressed using video codecs, as illustrated in Fig. 4. In the later case, a dedicated codec may be designed to efficiently handle the compression of the data, as shown in Fig. 5. In both cases, codec performance is evaluated using multiview video as comformance points. Since rendered viewpoint video is generally not included in the input multiview videos, for the convenience of performance evaluation, the multiview videos need to be split into test views that are not included in the input multiview videos. 3.2 Point Cloud Perspective
[0077] Like point cloud sequences, 3D / 4D Gaussian Splatting data may be seen as a collection of xyz positions in time and associated attributes. In point clouds, these attributes represent the point properties at precise xyz locations. In contrast to point clouds, Gaussian Splatting data contains attributes that represent Gaussians centered at xyz positions. This format facilitates view-dependent rendering but requires additional attribute information per Gaussian point, including 3 scales, 4 rotations, 48 SH coefficients (16 per color channel) , and 1 opacity attribute.
[0078] From this perspective, the similarities between point clouds and the Gaussian representation raise questions about how well current point cloud compression approaches, like G-PCC, V-PCC or AI-based methods can compress Gaussian Splatting data.
[0079] Fig. 6 illustrates main modules highlighted so far in the approaches discussed in the WG7 MPEG 147 meeting. 4. Problems
[0080] The existing designs for compressing 3D Gaussian Splatting (3DGS) models using G-PCC have the following problems: 1. In the current design, G-PCC's current attribute support is limited to conventional point cloud attributes such as Colour, Reflectance, Opacity, Normal vector, Material identifier, Frame number, and frame index. However, it lacks explicit support and definitions for 3DGS-specific attributes including scale, rotation, and spherical harmonic coefficients. This limitation impedes efficient compression of 3DGS data using G-PCC. 2. In the current design, the current attribute compression algorithms in G-PCC are not optimized for 3DGS model attributes, resulting in suboptimal compression efficiency for these specialized data types. 3. G-PCC only supports integer precision data compression for both geometry and attribute information. However, 3DGS models contain floating-point precision data for both geometry and attributes, neces-sitating pre-processing before encoding and post-processing after decoding to accommodate G-PCC's compression algorithms. 5. Detailed solutions
[0081] To solve the above problems and some other problems not mentioned, methods as summarized below are disclosed. The embodiments should be considered as examples to explain the general concepts and should not be interpreted in a narrow way. Furthermore, these embodiments may be applied individually or combined in any manner. 1) Before using G-PCC to compress Gaussian splatting, since G-PCC can only handle integer precision, pre-processing may be used to quantize the floating-point values of geometry and attribute information to inte-gers. a. In one example, the geometry information may be positions. b. In one example, the attribute information may be scales, rotations, spherical harmonic (SH) coeffi-cients, opacities. c. In another example, the SH coefficients may need to be converted to color representations before quantization and compression using G-PCC. i. In one example, the color representation may be RGB. ii. In one example, the color representation may be YUV. d. In one example, the Gaussian Splatting models may be 2D / 3D / 4D Gaussian Splatting models. 2) After encoding and decoding with G-PCC, post-processing is used to dequantize these integers back to their original floating-point precision for the geometry and attribute information of Gaussian Splatting models. a. In one example, the geometry information may be positions. b. In one example, the attribute information may be scales, rotations, spherical harmonic (SH) coeffi-cients, opacities. c. In another example, the SH coefficients may need to be converted back from color representations to their original form after dequantization. i. In one example, the color representation may be RGB. ii. In one example, the color representation may be YUV. d. In one example, the Gaussian Splatting models may be 2D / 3D / 4D Gaussian Splatting models. 3) The geometry information of Gaussian Splatting models may be compressed using G-PCC intra geometry compression methods. a. In one example, the geometry information of Gaussian Splatting models may be compressed using Octree-based Geometry Coding. b. In another example, Predictive Geometry Coding may be used for compressing the geometry infor-mation of Gaussian Splatting models. c. In another example, Trisoup Geometry Coding may be applied to compress the geometry infor-mation of Gaussian Splatting models. d. In one example, the geometry information of Gaussian Splatting models may be position. e. In one example, the Gaussian Splatting models may be 2D / 3D / 4D Gaussian Splatting models. 4) The attribute information of Gaussian Splatting models may be compressed using G-PCC intra attribute compression methods. a. In one example, the attribute information of Gaussian Splatting models may be compressed using RAHT (Region-Adaptive Hierarchical Transform) . b. In one example, the attribute information of Gaussian Splatting models may be compressed using Lifting Transform. c. In one example, the attribute information of Gaussian Splatting models may be compressed using Predicting Transform. d. In one example, attribute information of Gaussian Splatting models may be scales, rotations, spher-ical harmonic (SH) coefficients and opacities. e. In one example, the Gaussian Splatting models may be 2D / 3D / 4D Gaussian Splatting models. 5) The geometry information of Gaussian Splatting models may be compressed using G-PCC inter-frame ge-ometry compression methods. a. In one example, the geometry information of Gaussian Splatting models may be compressed using inter-frame Octree-based Geometry Coding. b. In another example, inter-frame Predictive Geometry Coding may be used for compressing the ge-ometry information of Gaussian Splatting models. c. In another example, inter-frame Trisoup Geometry Coding may be applied to compress the geom-etry information of Gaussian Splatting models. d. In one example, the geometry information of Gaussian Splatting models may be position. e. In one example, the Gaussian Splatting models may be 2D / 3D / 4D Gaussian Splatting models. 6) The attribute information of Gaussian Splatting models may be compressed using G-PCC inter-frame at-tribute compression methods. a. In one example, the attribute information of Gaussian Splatting models may be compressed using inter-frame RAHT (Region-Adaptive Hierarchical Transform) . b. In one example, the attribute information of Gaussian Splatting models may be compressed using inter-frame Lifting Transform. c. In one example, the attribute information of Gaussian Splatting models may be compressed using inter-frame Predicting Transform. d. In one example, attribute information of Gaussian Splatting models may be scales, rotations, spher-ical harmonic (SH) coefficients and opacities. e. In one example, the Gaussian Splatting models may be 2D / 3D / 4D Gaussian Splatting models. 7) Different attributes of Gaussian Splatting models may use different G-PCC attribute compression methods or same G-PCC attribute compression method. a. In one example, different attributes of the Gaussian Splatting model, such as scales, rotations, SH coefficients, and opacities, can be compressed using different G-PCC attribute compression meth-ods and inter-frame or intra-frame techniques. b. In another example, scales and rotations can be compressed using RAHT with intra-frame encoding, while SH coefficients and opacities are compressed using Predictive Transform with inter-frame encoding. c. In another example, all attributes can be compressed using the same method, such as Lifting Trans-form, with either inter-frame or intra-frame techniques. d. In one example, the Gaussian Splatting models may be 2D / 3D / 4D Gaussian Splatting models. 8) Different components of attribute of Gaussian Splatting models may use different G-PCC attribute com-pression methods or same G-PCC attribute compression method. a. In one example, the different components of a single attribute, such as the x, y, and z components of the scale attribute, can be compressed using different G-PCC attribute compression methods, such as RAHT for the x component and Predictive Transform for the y and z components. b. In another example, the components of the SH coefficients can be compressed using a combination of intra-frame and inter-frame methods, with intra-frame Lifting Transform for lower-order coeffi-cients and inter-frame Predictive Transform for higher-order coefficients. c. In one example, the Gaussian Splatting models may be 2D / 3D / 4D Gaussian Splatting models. 9) SH coefficients compression of Gaussian Splatting models may use different processing methods. a. In one example, SH coefficients can be directly quantized and then compressed using G-PCC at-tribute compression method. b. In another example, SH coefficients can be transformed into color representation, and then com-pressed using G-PCC's color compression methods. i. In one example, the color representation may be RGB. ii. In one example, the color representation may be YUV. c. In one example, the Gaussian Splatting models may be 2D / 3D / 4D Gaussian Splatting models. 10) At least one attribute labels may be added for identifying Gaussian Splatting model attributes. a. In one example, a label may be added to identify the SH coefficients attribute. b. In another example, a label may be added to identify the scales attribute. c. In another example, a label may be added to identify the rotations attribute. d. In another example, a label may be added to identify the opacities attribute. e. In one example, these added labels may be signalled in a code unit. i. In one example, the code unit may SPS / APS / GPS / Attribute Header / Geometry Header, etc. f. In one example, the Gaussian Splatting models may be 2D / 3D / 4D Gaussian Splatting models. 11) The Gaussian Splatting models may be divided into smaller units to enhance parallel encoding and prevent error propagation. a. In one example, the unit may be tile. b. In one example, the unit may be slice. 12) Whether to and / or how to apply a method disclosed above may be signaled from encoder to decoder in a bitstream / frame / tile / slice / octree / etc. 13) Whether to and / or how to apply the disclosed methods above may be dependent on coded information, such as dimensions, colour format, colour component, slice / picture type. 6. Embodiments
[0082] An example of the coding flow 700 for compressing Gaussian Splatting model using geometry-based point cloud compression (G-PCC) is depicted in Fig. 7. The coding flow 700 may be implemented by an electronic device or a computing device such as a point cloud coding device or video coding device.
[0083] As shown, at block 710, a pre-processing is applied to a Gaussian splatting model. At block 720, a G-PCC encoding is applied to the pre-processed Gaussian splatting model, to obtain a G-PCC bitstream. At block 730, a G-PCC decoding is applied to the G-PCC bitstream to obtain decoded information such as decoded Gaussian splatting model. At block 740, a post-processing is applied to the decoded Gaussian splatting model to obtain a reconstructed Gaussian Splatting model.
[0084] Further embodiments will be described with reference to Fig. 8 to Fig. 10 below. Fig. 8 illustrates a flowchart of a method 800 for video processing in accordance with embodiments of the present disclosure. The method 800 is implemented during a conversion between current Gaussian splatting (GS) model of a video and a bitstream of the video.
[0085] At block 810, for a conversion between a current Gaussian splatting (GS) model of a video and a bitstream of the video, geometry information and attribute information of the current GS model are obtained. The current GS model is represented by positions in a space with the attribute information associated with the positions. The geometry information includes the positions. In some embodiments, the conversion includes encoding the current GS model into the bitstream. Alternatively, or in addition, in some embodiments, the conversion includes decoding the current GS model from the bitstream.
[0086] At block 820, the geometry information and the attribute information are pre-processed to quantize floating-point values of the geometry information and the attribute information to integer values.
[0087] At block 830, the conversion is performed by applying at least one coding tool (or at least one coding mode) for G-PCC to the pre-processed geometry information and attribute information. The at least one coding tool (or at least one coding mode) for G-PCC may refer to any suitable coding tool or coding mode or coding algorithm used in G-PCC. Partial coding tool (s) or all coding tools for G-PCC may be applied for GS model coding. In the following description, several coding tools or coding modes for G-PCC will be described for the purpose of illustration, without suggesting any limitation to applicable G-PCC coding tool or coding mode. Embodiments of the present disclosure are not limited here.
[0088] The method 800 enables pre-processing the floating-point precision data for both geometry and attribute of the GS model before coding to accommodate G-PCC’s compression algorithms.
[0089] In some embodiments, the attribute information comprises at least one of: scale information, rotation information, spherical harmonic (SH) coefficients or opacity information of the current GS model.
[0090] In some embodiments, the attribute information comprises three scale components (or channels) , 4 rotation components (or channels) , 48 SH coefficient components (or channels) and 1 opacity attribute component (or channel) .
[0091] In some embodiments, the attribute information comprises SH coefficients of the current GS model, and the SH coefficients are converted to color representations before being quantized and compressed using the at least one coding tool for G-PCC. By way of example, the color representations may be RGB representations, or YUV representations. In an example, the SH coefficients may include data of 3 DC channels and 45 AC channels. Every three channels of the 3DC channels and 45 AC channels may be grouped and then converted into RGB format or YUV format, and the like.
[0092] In some embodiments, the current GS model comprises one of: a 2D GS model, a 3D GS model, a 4D GS model, or the like. The 4D GS model may correspond to 3D in space and one dimension in time domain.
[0093] In some embodiments, the method 800 further comprises: after coding the pre-processed geometry information and attribute information with the at least one coding tool for G-PCC, post-processing the integer values of the coded geometry information and coded attribute information to dequantize the integer values to floating-point precisions. In an example, the attribute information comprises SH coefficients of the current GS model, and the SH coefficients are converted from color representations to an original form after dequantization. In this manner, the pre-processing before encoding and the post-processing after decoding may be applied to accommodate G-PCC’s compression algorithms.
[0094] In some embodiments, the at least one coding tool for G-PCC comprises at least one of: a G-PCC intra geometry compression, a G-PCC intra attribute compression, a G-PCC inter-frame geometry compression, or a G-PCC inter-frame attribute compression.
[0095] In some embodiments, the current GS model comprises a plurality of units for parallel coding. For example, the plurality of units may be a plurality of tiles, a plurality of slices, or any other suitable coding units.
[0096] According to further embodiments of the present disclosure, a non-transitory computer-readable recording medium is provided. The non-transitory computer-readable recording medium stores a bitstream of a video which is generated by a method performed by a video processing apparatus. In the method, geometry information and attribute information of a current Gaussian splatting (GS) model of the video are obtained. The current GS model is represented by positions in a space with the attribute information associated with the positions. The geometry information includes the positions. The geometry information and the attribute information are pre-processed to quantize floating-point values of the geometry information and the attribute information to integer values. The bitstream is generated by applying at least one coding tool for geometry-based point cloud compression (G-PCC) to the pre-processed geometry information and attribute information.
[0097] According to still further embodiments of the present disclosure, a method for storing bitstream of a video is provided. In the method, geometry information and attribute information of a current Gaussian splatting (GS) model of the video are obtained. The current GS model is represented by positions in a space with the attribute information associated with the positions. The geometry information includes the positions. The geometry information and the attribute information are pre-processed to quantize floating-point values of the geometry information and the attribute information to integer values. The bitstream is generated by applying at least one coding tool for geometry-based point cloud compression (G-PCC) to the pre-processed geometry information and attribute information. The bitstream is stored in a non-transitory computer-readable recording medium.
[0098] Fig. 9 illustrates a flowchart of a method 900 for video processing in accordance with embodiments of the present disclosure. The method 900 is implemented during a conversion between current Gaussian splatting (GS) model of a video and a bitstream of the video.
[0099] At block 910, for a conversion between a current Gaussian splatting (GS) model of a video and a bitstream of the video, geometry information and attribute information of the current GS model are obtained. The current GS model is represented by positions in a space with the attribute information associated with the positions. The geometry information includes the positions. The attribute information comprises at least one of: scale information, rotation information, spherical harmonic (SH) coefficients or opacity information of the current GS model.
[0100] At block 920, at least one coding tool for geometry-based point cloud compression (G-PCC) is applied to at least one of the geometry information or the attribute information.
[0101] At block 930, the conversion is performed based on the applying. In some embodiments, the conversion includes encoding the current GS model into the bitstream. Alternatively, or in addition, in some embodiments, the conversion includes decoding the current GS model from the bitstream.
[0102] The method 900 enables supporting of different types of attribute information of GS model, such as SH coefficients, scale, rotation and opacity.
[0103] In some embodiments, the current GS model may be a 2D GS model, a 3D GS model, a 4D GS model, or the like.
[0104] In some embodiments, the at least one coding tool for G-PCC may be a G-PCC intra geometry compression tool, and the geometry information of the current GS model is compressed using the G-PCC intra geometry compression tool. In some embodiments, the G-PCC intra geometry compression tool may be at least one of: an octree-based geometry coding, a predictive geometry coding, or a Trisoup geometry coding.
[0105] In some embodiments, the at least one coding tool for G-PCC may be a G-PCC intra attribute compression tool, and the attribute information of the current GS model is compressed using the G-PCC intra attribute compression tool. In some embodiments, the G-PCC intra attribute compression tool may be at least one of: a region adaptive hierarchical transform (RAHT) based attribute coding, a lifting transform based attribute coding, or a predicting transform based attribute coding.
[0106] In some embodiments, the at least one coding tool for G-PCC may be a G-PCC inter-frame geometry compression tool, and the geometry information of the current GS model is compressed using the G-PCC inter-frame geometry compression tool. In some embodiments, the G-PCC inter-frame geometry compression tool may be at least one of: an inter-frame octree-based geometry coding, an inter-frame predictive geometry coding, or an inter-frame Trisoup geometry coding.
[0107] In some embodiments, the at least one coding tool for G-PCC may be a G-PCC inter-frame attribute compression tool, and the attribute information of the current GS model is compressed using the G-PCC inter-frame attribute compression tool. In some embodiments, the G-PCC inter-frame attribute compression tool may be at least one of: an inter-frame region adaptive hierarchical transform (RAHT) based attribute coding, an inter-frame lifting transform based attribute coding, or an inter-frame predicting transform based attribute coding.
[0108] In some embodiments, a first attribute of the current GS model may be compressed by a first G-PCC attribute compression tool, and a second attribute of the current GS model may be compressed by a second G-PCC attribute compression tool different from the first G-PCC attribute compression tool. In some embodiments, the first attribute and the second attribute are compressed further using different inter-frame or intra-frame techniques.
[0109] In some embodiments, the first attribute may be at least one of the scale information or the rotation information, and the first G-PCC attribute compression tool comprises RAHT based attribute coding with intra-frame coding. The second attribute may be at least one of the SH coefficients or the opacity information, and the second G-PCC attribute compression tool comprises a predictive transform based attribute coding with inter-frame coding.
[0110] In some embodiments, all attributes of the current GS model are compressed with a same G-PCC compression tool. For example, the same G-PCC compression tool comprising a lifting transform based attribute coding with inter-frame or intra-frame technique.
[0111] In some embodiments, a first component of the attribute information of the current GS model is compressed by a third G-PCC attribute compression tool, and a second component of the attribute information of the current GS model is compressed by a fourth G-PCC attribute compression tool different from the third G-PCC attribute compression tool.
[0112] In some embodiments, the first component comprises a x component (corresponding to x axis in a 3D space) of the attribute information, and the second component comprises at least one of y component (corresponding to y axis in the 3D space) or z component (corresponding to z axis in the 3D space) of the attribute information, the third G-PCC attribute compression tool comprising an RAHT based attribute coding, and the fourth G-PCC attribute compression tool comprising a predictive transform based attribute coding.
[0113] In some embodiments, components of the SH coefficients are compressed using a combination of intra-frame coding tool and inter-frame coding tool, an intra-frame lifting transform based attribute coding being used for low-order coefficients of the SH coefficients and an inter-frame predictive transform based attribute coding being used for higher-order coefficients of the SH coefficients.
[0114] In some embodiments, a first component and a second component of the attribute information of the current GS model are compressed by a same G-PCC attribute compression tool.
[0115] In some embodiments, the SH coefficients of the current GS model use a plurality of processing approaches.
[0116] In some embodiments, the SH coefficients are directly quantized and then compressed using at least one G-PCC attribute compression tool.
[0117] In some embodiments, the SH coefficients are transformed into color representations and then compressed using a G-PCC color compression tool. For example, the color representations may be RGB representations or YUV representations.
[0118] In some embodiments, at least one attribute label is indicated to identify at least one attribute of the current GS model. By way of example, the at least one label may include one or more of the following: a first label identifying the SH coefficients, a second label identifying the scale information, a third label identifying the rotation information, or a fourth label identifying the opacity information.
[0119] In some embodiments, the at least one label may be included in a code unit in the bitstream. For example, the code unit may be a sequence parameter set (SPS) , an attribute parameter set (APS) , a geometry parameter set (GPS) , an attribute header, or a geometry header.
[0120] In some embodiments, the current GS model may include a plurality of units for parallel coding. For example, the plurality of units may be a plurality of tiles, a plurality of slices, or the like.
[0121] According to further embodiments of the present disclosure, a non-transitory computer-readable recording medium is provided. The non-transitory computer-readable recording medium stores a bitstream of a video which is generated by a method performed by a video apparatus. In the method, geometry information and attribute information of a current Gaussian splatting (GS) model of the video are obtained. The current GS model is represented by positions in a space with the attribute information associated with the positions. The geometry information includes the positions. The attribute information comprises at least one of: scale information, rotation information, spherical harmonic (SH) coefficients or opacity information of the current GS model. At least one coding tool for G-PCC is applied to at least one of the geometry information or the attribute information. The bitstream is generated based on the applying.
[0122] According to still further embodiments of the present disclosure, a method for storing bitstream of a video is provided. In the method, geometry information and attribute information of a current Gaussian splatting (GS) model of the video are obtained. The current GS model is represented by positions in a space with the attribute information associated with the positions. The geometry information includes the positions. The attribute information comprises at least one of: scale information, rotation information, spherical harmonic (SH) coefficients or opacity information of the current GS model. At least one coding tool for G-PCC is applied to at least one of the geometry information or the attribute information. The bitstream is generated based on the applying. The bitstream is stored in a non-transitory computer-readable recording medium.
[0123] It is to be understood that the method 800 and / or the method 900 may be applied separately, or in any combination. For example, the coding flow 700 in Fig. 7 illustrates a combination of the method 800 and the method 900. Specifically, at 710, the pre-processing described with respect to the method 800 may be applied. At 740, the post-processing described with respect to the method 800 may be applied. At 720 and / or 730, the at least one coding tool for G-PCC as described with respect to the method 900 may be applied. Embodiments of the present disclosure are not limited here.
[0124] In some embodiments, whether to and / or how to apply the method 800 and / or the method 900 may be indicated in one or more of the following: the bitstream, or a frame, a tile, a slice, or an octree in the bitstream.
[0125] In some embodiments, whether to and / or how to apply the the method 800 and / or the method 900 may be based on coded information. For example, the coded information may include one or more of the following: dimension information, a color format, a color component, a slice type or a picture type.
[0126] Implementations of the present disclosure can be described in view of the following clauses, the features of which can be combined in any reasonable manner.
[0127] Clause 1. A method for video processing, comprising: obtaining, for a conversion between a current Gaussian splatting (GS) model of a video and a bitstream of the video, geometry information and attribute information of the current GS model, the current GS model being represented by positions in a space with the attribute information associated with the positions, the geometry information comprising the positions; pre-processing the geometry information and the attribute information to quantize floating-point values of the geometry information and the attribute information to integer values; and performing the conversion by applying at least one coding tool for geometry-based point cloud compression (G-PCC) to the pre-processed geometry information and attribute information.
[0128] Clause 2. The method of clause 1, wherein the attribute information comprises at least one of: scale information, rotation information, spherical harmonic (SH) coefficients or opacity information of the current GS model.
[0129] Clause 3. The method of clause 2, wherein the attribute information comprises three scale components, 4 rotation components, 48 SH coefficient components and 1 opacity attribute component.
[0130] Clause 4. The method of clause 2 or 3, wherein the attribute information comprises SH coefficients of the current GS model, and the SH coefficients are converted to color representations before being quantized and compressed using the at least one coding tool for G-PCC.
[0131] Clause 5. The method of clause 4, wherein the color representations comprise one of: RGB representations, or YUV representations.
[0132] Clause 6. The method of any of clauses 1 to 5, wherein the current GS model comprises one of: a 2D GS model, a 3D GS model, or a 4D GS model.
[0133] Clause 7. The method of any of clauses 1 to 6, further comprising: after coding the pre-processed geometry information and attribute information with the at least one coding tool for G-PCC, post-processing the integer values of the coded geometry information and coded attribute information to dequantize the integer values to floating-point precisions.
[0134] Clause 8. The method of clause 7, wherein the attribute information comprises SH coefficients of the current GS model, and the SH coefficients are converted from color representations to an original form after dequantization.
[0135] Clause 9. The method of any of clauses 1 to 8, wherein the at least one coding tool for G-PCC comprises at least one of: a G-PCC intra geometry compression, a G-PCC intra attribute compression, a G-PCC inter-frame geometry compression, or a G-PCC inter-frame attribute compression.
[0136] Clause 10. The method of any of clauses 1 to 9, wherein the current GS model comprises a plurality of units for parallel coding.
[0137] Clause 11. The method of clause 10, wherein the plurality of units comprises a plurality of tiles or a plurality of slices.
[0138] Clause 12. A method for video processing, comprising: obtaining, for a conversion between a current Gaussian splatting (GS) model of a video and a bitstream of the video, geometry information and attribute information of the current GS model, the current GS model being represented by positions in a space with the attribute information associated with the positions, the geometry information comprising the positions; applying at least one coding tool for geometry-based point cloud compression (G-PCC) to at least one of the geometry information or the attribute information; and performing the conversion based on the applying, wherein the attribute information comprises at least one of: scale information, rotation information, spherical harmonic (SH) coefficients or opacity information of the current GS model.
[0139] Clause 13. The method of clause 12, wherein the current GS model comprises one of: a 2D GS model, a 3D GS model, or a 4D GS model.
[0140] Clause 14. The method of clause 12 or 13, wherein the at least one coding tool for G-PCC comprises a G-PCC intra geometry compression tool, and the geometry information of the current GS model is compressed using the G-PCC intra geometry compression tool.
[0141] Clause 15. The method of clause 14, wherein the G-PCC intra geometry compression tool comprises at least one of: an octree-based geometry coding, a predictive geometry coding, or a Trisoup geometry coding.
[0142] Clause 16. The method of any of clauses 12 to 15, wherein the at least one coding tool for G-PCC comprises a G-PCC intra attribute compression tool, and the attribute information of the current GS model is compressed using the G-PCC intra attribute compression tool.
[0143] Clause 17. The method of clause 16, wherein the G-PCC intra attribute compression tool comprises at least one of: a region adaptive hierarchical transform (RAHT) based attribute coding, a lifting transform based attribute coding, or a predicting transform based attribute coding.
[0144] Clause 18. The method of any of clauses 12 to 17, wherein the at least one coding tool for G-PCC comprises a G-PCC inter-frame geometry compression tool, and the geometry information of the current GS model is compressed using the G-PCC inter-frame geometry compression tool.
[0145] Clause 19. The method of clause 18, wherein the G-PCC inter-frame geometry compression tool comprises at least one of: an inter-frame octree-based geometry coding, an inter-frame predictive geometry coding, or an inter-frame Trisoup geometry coding.
[0146] Clause 20. The method of any of clauses 12 to 19, wherein the at least one coding tool for G-PCC comprises a G-PCC inter-frame attribute compression tool, and the attribute information of the current GS model is compressed using the G-PCC inter-frame attribute compression tool.
[0147] Clause 21. The method of clause 20, wherein the G-PCC inter-frame attribute compression tool comprises at least one of: an inter-frame region adaptive hierarchical transform (RAHT) based attribute coding, an inter-frame lifting transform based attribute coding, or an inter-frame predicting transform based attribute coding.
[0148] Clause 22. The method of any of clauses 12 to 21, wherein a first attribute of the current GS model is compressed by a first G-PCC attribute compression tool, and a second attribute of the current GS model is compressed by a second G-PCC attribute compression tool different from the first G-PCC attribute compression tool.
[0149] Clause 23. The method of clause 22, wherein the first attribute and the second attribute are compressed further using different inter-frame or intra-frame techniques.
[0150] Clause 24. The method of clause 22 or 23, wherein the first attribute comprises at least one of the scale information or the rotation information, and the first G-PCC attribute compression tool comprises RAHT based attribute coding with intra-frame coding, and wherein the second attribute comprises at least one of the SH coefficients or the opacity information, and the second G-PCC attribute compression tool comprises a predictive transform based attribute coding with inter-frame coding.
[0151] Clause 25. The method of any of clauses 12 to 21, wherein all attributes of the current GS model are compressed with a same G-PCC compression tool.
[0152] Clause 26. The method of clause 25, wherein the same G-PCC compression tool comprising a lifting transform based attribute coding with inter-frame or intra-frame technique.
[0153] Clause 27. The method of any of clauses 12 to 26, wherein a first component of the attribute information of the current GS model is compressed by a third G-PCC attribute compression tool, and a second component of the attribute information of the current GS model is compressed by a fourth G-PCC attribute compression tool different from the third G-PCC attribute compression tool.
[0154] Clause 28. The method of clause 27, wherein the first component comprises a x component of the attribute information, and the second component comprises at least one of y component or z component of the attribute information, the third G-PCC attribute compression tool comprising an RAHT based attribute coding, and the fourth G-PCC attribute compression tool comprising a predictive transform based attribute coding.
[0155] Clause 29. The method of clause 27, wherein components of the SH coefficients are compressed using a combination of intra-frame coding tool and inter-frame coding tool, an intra-frame lifting transform based attribute coding being used for low-order coefficients of the SH coefficients and an inter-frame predictive transform based attribute coding being used for higher-order coefficients of the SH coefficients.
[0156] Clause 30. The method of any of clauses 12 to 26, wherein a first component and a second component of the attribute information of the current GS model are compressed by a same G-PCC attribute compression tool.
[0157] Clause 31. The method of any of clauses 12 to 30, wherein the SH coefficients of the current GS model use a plurality of processing approaches.
[0158] Clause 32. The method of clause 31, wherein the SH coefficients are directly quantized and then compressed using at least one G-PCC attribute compression tool.
[0159] Clause 33. The method of clause 31, wherein the SH coefficients are transformed into color representations and then compressed using a G-PCC color compression tool.
[0160] Clause 34. The method of clause 33, wherein the color representations comprise one of: RGB representations or YUV representations.
[0161] Clause 35. The method of any of clauses 12 to 34, wherein at least one attribute label is indicated to identify at least one attribute of the current GS model.
[0162] Clause 36. The method of clause 35, wherein the at least one label comprises at least one of: a first label identifying the SH coefficients, a second label identifying the scale information, a third label identifying the rotation information, or a fourth label identifying the opacity information.
[0163] Clause 37. The method of clause 35 or 36, wherein the at least one label is included in a code unit in the bitstream.
[0164] Clause 38. The method of clause 37, wherein the code unit comprises one of: a sequence parameter set (SPS) , an attribute parameter set (APS) , a geometry parameter set (GPS) , an attribute header, or a geometry header.
[0165] Clause 39. The method of any of clauses 23 to 38, wherein the current GS model comprises a plurality of units for parallel coding.
[0166] Clause 40. The method of clause 39, wherein the plurality of units comprises a plurality of tiles or a plurality of slices.
[0167] Clause 41. The method of any of clauses 1 to 40, wherein whether to apply the method and / or how to apply the method is indicated in at least one of: the bitstream, or a frame, a tile, a slice, or an octree in the bitstream.
[0168] Clause 42. The method of any of clauses 1 to 40, wherein whether to apply the method and / or how to apply the method is based on coded information, the coded information comprising at least one of: dimension information, a color format, a color component, a slice type or a picture type.
[0169] Clause 43. The method of any of clauses 1-42, wherein the conversion includes encoding the current GS model into the bitstream.
[0170] Clause 44. The method of any of clauses 1-42, wherein the conversion includes decoding the current GS model from the bitstream.
[0171] Clause 45. An apparatus for video processing comprising a processor and a non-transitory memory with instructions thereon, wherein the instructions upon execution by the processor, cause the processor to perform a method in accordance with any of clauses 1-44.
[0172] Clause 46. A non-transitory computer-readable storage medium storing instructions that cause a processor to perform a method in accordance with any of clauses 1-44.
[0173] Clause 47. A non-transitory computer-readable recording medium storing a bitstream of a video which is generated by a method performed by a video processing apparatus, wherein the method comprises: obtaining geometry information and attribute information of a current Gaussian splatting (GS) model of the video, the current GS model being represented by positions in a space with the attribute information associated with the positions, the geometry information comprising the positions; pre-processing the geometry information and the attribute information to quantize floating-point values of the geometry information and the attribute information to integer values; and generating the bitstream by applying at least one coding tool for geometry-based point cloud compression (G-PCC) to the pre-processed geometry information and attribute information.
[0174] Clause 48. A method for storing a bitstream of a video, comprising: obtaining geometry information and attribute information of a current Gaussian splatting (GS) model of the video, the current GS model being represented by positions in a space with the attribute information associated with the positions, the geometry information comprising the positions; pre-processing the geometry information and the attribute information to quantize floating-point values of the geometry information and the attribute information to integer values; generating the bitstream by applying at least one coding tool for geometry-based point cloud compression (G-PCC) to the pre-processed geometry information and attribute information; and storing the bitstream in a non-transitory computer-readable recording medium.
[0175] Clause 49. A non-transitory computer-readable recording medium storing a bitstream of a video which is generated by a method performed by a video processing apparatus, wherein the method comprises: obtaining geometry information and attribute information of a current Gaussian splatting (GS) model of the video, the current GS model being represented by positions in a space with the attribute information associated with the positions, the geometry information comprising the positions, wherein the attribute information comprises at least one of: scale information, rotation information, spherical harmonic (SH) coefficients or opacity information of the current GS model; applying at least one coding tool for geometry-based point cloud compression (G-PCC) to at least one of the geometry information or the attribute information; and generating the bitstream based on the applying.
[0176] Clause 50. A method for storing a bitstream of a video, comprising: obtaining geometry information and attribute information of a current Gaussian splatting (GS) model of the video, the current GS model being represented by positions in a space with the attribute information associated with the positions, the geometry information comprising the positions, wherein the attribute information comprises at least one of: scale information, rotation information, spherical harmonic (SH) coefficients or opacity information of the current GS model; applying at least one coding tool for geometry-based point cloud compression (G-PCC) to at least one of the geometry information or the attribute information; generating the bitstream based on the applying; and storing the bitstream in a non-transitory computer-readable recording medium. Example Device
[0177] Fig. 10 illustrates a block diagram of a computing device 1000 in which various embodiments of the present disclosure can be implemented. The computing device 1000 may be implemented as or included in the source device 110 (or the G-PCC encoder 116 or 200) or the destination device 120 (or the G-PCC decoder 126 or 300) .
[0178] It would be appreciated that the computing device 1000 shown in Fig. 10 is merely for purpose of illustration, without suggesting any limitation to the functions and scopes of the embodiments of the present disclosure in any manner.
[0179] As shown in Fig. 10, the computing device 1000 includes a general-purpose computing device 1000. The computing device 1000 may at least comprise one or more processors or processing units 1010, a memory 1020, a storage unit 1030, one or more communication units 1040, one or more input devices 1050, and one or more output devices 1060.
[0180] In some embodiments, the computing device 1000 may be implemented as any user terminal or server terminal having the computing capability. The server terminal may be a server, a large-scale computing device or the like that is provided by a service provider. The user terminal may for example be any type of mobile terminal, fixed terminal, or portable terminal, including a mobile phone, station, unit, device, multimedia computer, multimedia tablet, Internet node, communicator, desktop computer, laptop computer, notebook computer, netbook computer, tablet computer, personal communication system (PCS) device, personal navigation device, personal digital assistant (PDA) , audio / video player, digital camera / video camera, positioning device, television receiver, radio broadcast receiver, E-book device, gaming device, or any combination thereof, including the accessories and peripherals of these devices, or any combination thereof. It would be contemplated that the computing device 1000 can support any type of interface to a user (such as “wearable” circuitry and the like) .
[0181] The processing unit 1010 may be a physical or virtual processor and can implement various processes based on programs stored in the memory 1020. In a multi-processor system, multiple processing units execute computer executable instructions in parallel so as to improve the parallel processing capability of the computing device 1000. The processing unit 1010 may also be referred to as a central processing unit (CPU) , a microprocessor, a controller or a microcontroller.
[0182] The computing device 1000 typically includes various computer storage medium. Such medium can be any medium accessible by the computing device 1000, including, but not limited to, volatile and non-volatile medium, or detachable and non-detachable medium. The memory 1020 can be a volatile memory (for example, a register, cache, Random Access Memory (RAM) ) , a non-volatile memory (such as a Read-Only Memory (ROM) , Electrically Erasable Programmable Read-Only Memory (EEPROM) , or a flash memory) , or any combination thereof. The storage unit 1030 may be any detachable or non-detachable medium and may include a machine-readable medium such as a memory, flash memory drive, magnetic disk or another other media, which can be used for storing information and / or data and can be accessed in the computing device 1000.
[0183] The computing device 1000 may further include additional detachable / non-detachable, volatile / non-volatile memory medium. Although not shown in Fig. 10, it is possible to provide a magnetic disk drive for reading from and / or writing into a detachable and non-volatile magnetic disk and an optical disk drive for reading from and / or writing into a detachable non-volatile optical disk. In such cases, each drive may be connected to a bus (not shown) via one or more data medium interfaces.
[0184] The communication unit 1040 communicates with a further computing device via the communication medium. In addition, the functions of the components in the computing device 1000 can be implemented by a single computing cluster or multiple computing machines that can communicate via communication connections. Therefore, the computing device 1000 can operate in a networked environment using a logical connection with one or more other servers, networked personal computers (PCs) or further general network nodes.
[0185] The input device 1050 may be one or more of a variety of input devices, such as a mouse, keyboard, tracking ball, voice-input device, and the like. The output device 1060 may be one or more of a variety of output devices, such as a display, loudspeaker, printer, and the like. By means of the communication unit 1040, the computing device 1000 can further communicate with one or more external devices (not shown) such as the storage devices and display device, with one or more devices enabling the user to interact with the computing device 1000, or any devices (such as a network card, a modem and the like) enabling the computing device 1000 to communicate with one or more other computing devices, if required. Such communication can be performed via input / output (I / O) interfaces (not shown) .
[0186] In some embodiments, instead of being integrated in a single device, some or all components of the computing device 1000 may also be arranged in cloud computing architecture. In the cloud computing architecture, the components may be provided remotely and work together to implement the functionalities described in the present disclosure. In some embodiments, cloud computing provides computing, software, data access and storage service, which will not require end users to be aware of the physical locations or configurations of the systems or hardware providing these services. In various embodiments, the cloud computing provides the services via a wide area network (such as Internet) using suitable protocols. For example, a cloud computing provider provides applications over the wide area network, which can be accessed through a web browser or any other computing components. The software or components of the cloud computing architecture and corresponding data may be stored on a server at a remote position. The computing resources in the cloud computing environment may be merged or distributed at locations in a remote data center. Cloud computing infrastructures may provide the services through a shared data center, though they behave as a single access point for the users. Therefore, the cloud computing architectures may be used to provide the components and functionalities described herein from a service provider at a remote location. Alternatively, they may be provided from a conventional server or installed directly or otherwise on a client device.
[0187] The computing device 1000 may be used to implement point cloud encoding / decoding in embodiments of the present disclosure. The memory 1020 may include one or more point cloud coding modules 1025 having one or more program instructions. These modules are accessible and executable by the processing unit 1010 to perform the functionalities of the various embodiments described herein.
[0188] In the example embodiments of performing point cloud encoding, the input device 1050 may receive point cloud data as an input 1070 to be encoded. The point cloud data may be processed, for example, by the point cloud coding module 1025, to generate an encoded bitstream. The encoded bitstream may be provided via the output device 1060 as an output 1080.
[0189] In the example embodiments of performing point cloud decoding, the input device 1050 may receive an encoded bitstream as the input 1070. The encoded bitstream may be processed, for example, by the point cloud coding module 1025, to generate decoded point cloud data. The decoded point cloud data may be provided via the output device 1060 as the output 1080.
[0190] While this disclosure has been particularly shown and described with references to example embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the present application as defined by the appended claims. Such variations are intended to be covered by the scope of this present application. As such, the foregoing description of embodiments of the present application is not intended to be limiting.
Claims
1.A method for video processing, comprising:obtaining, for a conversion between a current Gaussian splatting (GS) model of a video and a bitstream of the video, geometry information and attribute information of the current GS model, the current GS model being represented by positions in a space with the attribute information associated with the positions, the geometry information comprising the positions;pre-processing the geometry information and the attribute information to quantize floating-point values of the geometry information and the attribute information to integer values; andperforming the conversion by applying at least one coding tool for geometry-based point cloud compression (G-PCC) to the pre-processed geometry information and attribute information.2.The method of claim 1, wherein the attribute information comprises at least one of: scale information, rotation information, spherical harmonic (SH) coefficients or opacity information of the current GS model.3.The method of claim 2, wherein the attribute information comprises three scale components, 4 rotation components, 48 SH coefficient components and 1 opacity attribute component.4.The method of claim 2 or 3, wherein the attribute information comprises SH coefficients of the current GS model, and the SH coefficients are converted to color representations before being quantized and compressed using the at least one coding tool for G-PCC.5.The method of claim 4, wherein the color representations comprise one of: RGB representations, or YUV representations.6.The method of any of claims 1 to 5, wherein the current GS model comprises one of: a 2D GS model, a 3D GS model, or a 4D GS model.7.The method of any of claims 1 to 6, further comprising:after coding the pre-processed geometry information and attribute information with the at least one coding tool for G-PCC, post-processing the integer values of the coded geometry information and coded attribute information to dequantize the integer values to floating-point precisions.8.The method of claim 7, wherein the attribute information comprises SH coefficients of the current GS model, and the SH coefficients are converted from color representations to an original form after dequantization.9.The method of any of claims 1 to 8, wherein the at least one coding tool for G-PCC comprises at least one of:a G-PCC intra geometry compression,a G-PCC intra attribute compression,a G-PCC inter-frame geometry compression, ora G-PCC inter-frame attribute compression.10.The method of any of claims 1 to 9, wherein the current GS model comprises a plurality of units for parallel coding.11.The method of claim 10, wherein the plurality of units comprises a plurality of tiles or a plurality of slices.12.A method for video processing, comprising:obtaining, for a conversion between a current Gaussian splatting (GS) model of a video and a bitstream of the video, geometry information and attribute information of the current GS model, the current GS model being represented by positions in a space with the attribute information associated with the positions, the geometry information comprising the positions;applying at least one coding tool for geometry-based point cloud compression (G-PCC) to at least one of the geometry information or the attribute information; andperforming the conversion based on the applying,wherein the attribute information comprises at least one of: scale information, rotation information, spherical harmonic (SH) coefficients or opacity information of the current GS model.13.The method of claim 12, wherein the current GS model comprises one of: a 2D GS model, a 3D GS model, or a 4D GS model.14.The method of claim 12 or 13, wherein the at least one coding tool for G-PCC comprises a G-PCC intra geometry compression tool, and the geometry information of the current GS model is compressed using the G-PCC intra geometry compression tool.15.The method of claim 14, wherein the G-PCC intra geometry compression tool comprises at least one of:an octree-based geometry coding,a predictive geometry coding, ora Trisoup geometry coding.16.The method of any of claims 12 to 15, wherein the at least one coding tool for G-PCC comprises a G-PCC intra attribute compression tool, and the attribute information of the current GS model is compressed using the G-PCC intra attribute compression tool.17.The method of claim 16, wherein the G-PCC intra attribute compression tool comprises at least one of:a region adaptive hierarchical transform (RAHT) based attribute coding,a lifting transform based attribute coding, ora predicting transform based attribute coding.18.The method of any of claims 12 to 17, wherein the at least one coding tool for G-PCC comprises a G-PCC inter-frame geometry compression tool, and the geometry information of the current GS model is compressed using the G-PCC inter-frame geometry compression tool.19.The method of claim 18, wherein the G-PCC inter-frame geometry compression tool comprises at least one of:an inter-frame octree-based geometry coding,an inter-frame predictive geometry coding, oran inter-frame Trisoup geometry coding.20.The method of any of claims 12 to 19, wherein the at least one coding tool for G-PCC comprises a G-PCC inter-frame attribute compression tool, and the attribute information of the current GS model is compressed using the G-PCC inter-frame attribute compression tool.21.The method of claim 20, wherein the G-PCC inter-frame attribute compression tool comprises at least one of:an inter-frame region adaptive hierarchical transform (RAHT) based attribute coding,an inter-frame lifting transform based attribute coding, oran inter-frame predicting transform based attribute coding.22.The method of any of claims 12 to 21, wherein a first attribute of the current GS model is compressed by a first G-PCC attribute compression tool, and a second attribute of the current GS model is compressed by a second G-PCC attribute compression tool different from the first G-PCC attribute compression tool.23.The method of claim 22, wherein the first attribute and the second attribute are compressed further using different inter-frame or intra-frame techniques.24.The method of claim 22 or 23, wherein the first attribute comprises at least one of the scale information or the rotation information, and the first G-PCC attribute compression tool comprises RAHT based attribute coding with intra-frame coding, andwherein the second attribute comprises at least one of the SH coefficients or the opacity information, and the second G-PCC attribute compression tool comprises a predictive transform based attribute coding with inter-frame coding.25.The method of any of claims 12 to 21, wherein all attributes of the current GS model are compressed with a same G-PCC compression tool.26.The method of claim 25, wherein the same G-PCC compression tool comprising a lifting transform based attribute coding with inter-frame or intra-frame technique.27.The method of any of claims 12 to 26, wherein a first component of the attribute information of the current GS model is compressed by a third G-PCC attribute compression tool, and a second component of the attribute information of the current GS model is compressed by a fourth G-PCC attribute compression tool different from the third G-PCC attribute compression tool.28.The method of claim 27, wherein the first component comprises a x component of the attribute information, and the second component comprises at least one of y component or z component of the attribute information, the third G-PCC attribute compression tool comprising an RAHT based attribute coding, and the fourth G-PCC attribute compression tool comprising a predictive transform based attribute coding.29.The method of claim 27, wherein components of the SH coefficients are compressed using a combination of intra-frame coding tool and inter-frame coding tool, an intra-frame lifting transform based attribute coding being used for low-order coefficients of the SH coefficients and an inter-frame predictive transform based attribute coding being used for higher-order coefficients of the SH coefficients.30.The method of any of claims 12 to 26, wherein a first component and a second component of the attribute information of the current GS model are compressed by a same G-PCC attribute compression tool.31.The method of any of claims 12 to 30, wherein the SH coefficients of the current GS model use a plurality of processing approaches.32.The method of claim 31, wherein the SH coefficients are directly quantized and then compressed using at least one G-PCC attribute compression tool.33.The method of claim 31, wherein the SH coefficients are transformed into color representations and then compressed using a G-PCC color compression tool.34.The method of claim 33, wherein the color representations comprise one of: RGB representations or YUV representations.35.The method of any of claims 12 to 34, wherein at least one attribute label is indicated to identify at least one attribute of the current GS model.36.The method of claim 35, wherein the at least one label comprises at least one of:a first label identifying the SH coefficients,a second label identifying the scale information,a third label identifying the rotation information, ora fourth label identifying the opacity information.37.The method of claim 35 or 36, wherein the at least one label is included in a code unit in the bitstream.38.The method of claim 37, wherein the code unit comprises one of: a sequence parameter set (SPS) , an attribute parameter set (APS) , a geometry parameter set (GPS) , an attribute header, or a geometry header.39.The method of any of claims 12 to 38, wherein the current GS model comprises a plurality of units for parallel coding.40.The method of claim 39, wherein the plurality of units comprises a plurality of tiles or a plurality of slices.41.The method of any of claims 1 to 40, wherein whether to apply the method and / or how to apply the method is indicated in at least one of: the bitstream, or a frame, a tile, a slice, or an octree in the bitstream.42.The method of any of claims 1 to 40, wherein whether to apply the method and / or how to apply the method is based on coded information, the coded information comprising at least one of: dimension information, a color format, a color component, a slice type or a picture type.43.The method of any of claims 1-42, wherein the conversion includes encoding the current GS model into the bitstream.44.The method of any of claims 1-42, wherein the conversion includes decoding the current GS model from the bitstream.45.An apparatus for video processing comprising a processor and a non-transitory memory with instructions thereon, wherein the instructions upon execution by the processor, cause the processor to perform a method in accordance with any of claims 1-44.46.A non-transitory computer-readable storage medium storing instructions that cause a processor to perform a method in accordance with any of claims 1-44.47.A non-transitory computer-readable recording medium storing a bitstream of a video which is generated by a method performed by a video processing apparatus, wherein the method comprises:obtaining geometry information and attribute information of a current Gaussian splatting (GS) model of the video, the current GS model being represented by positions in a space with the attribute information associated with the positions, the geometry information comprising the positions;pre-processing the geometry information and the attribute information to quantize floating-point values of the geometry information and the attribute information to integer values; andgenerating the bitstream by applying at least one coding tool for geometry-based point cloud compression (G-PCC) to the pre-processed geometry information and attribute information.48.A method for storing a bitstream of a video, comprising:obtaining geometry information and attribute information of a current Gaussian splatting (GS) model of the video, the current GS model being represented by positions in a space with the attribute information associated with the positions, the geometry information comprising the positions;pre-processing the geometry information and the attribute information to quantize floating-point values of the geometry information and the attribute information to integer values;generating the bitstream by applying at least one coding tool for geometry-based point cloud compression (G-PCC) to the pre-processed geometry information and attribute information; andstoring the bitstream in a non-transitory computer-readable recording medium.49.A non-transitory computer-readable recording medium storing a bitstream of a video which is generated by a method performed by a video processing apparatus, wherein the method comprises:obtaining geometry information and attribute information of a current Gaussian splatting (GS) model of the video, the current GS model being represented by positions in a space with the attribute information associated with the positions, the geometry information comprising the positions, wherein the attribute information comprises at least one of: scale information, rotation information, spherical harmonic (SH) coefficients or opacity information of the current GS model;applying at least one coding tool for geometry-based point cloud compression (G-PCC) to at least one of the geometry information or the attribute information; andgenerating the bitstream based on the applying.50.A method for storing a bitstream of a video, comprising:obtaining geometry information and attribute information of a current Gaussian splatting (GS) model of the video, the current GS model being represented by positions in a space with the attribute information associated with the positions, the geometry information comprising the positions, wherein the attribute information comprises at least one of: scale information, rotation information, spherical harmonic (SH) coefficients or opacity information of the current GS model;applying at least one coding tool for geometry-based point cloud compression (G-PCC) to at least one of the geometry information or the attribute information;generating the bitstream based on the applying; andstoring the bitstream in a non-transitory computer-readable recording medium.