WO2022257971A1 - 点云编码处理方法、点云解码处理方法及相关设备 - Google Patents

点云编码处理方法、点云解码处理方法及相关设备 Download PDF

Info

Publication number
WO2022257971A1
WO2022257971A1 PCT/CN2022/097635 CN2022097635W WO2022257971A1 WO 2022257971 A1 WO2022257971 A1 WO 2022257971A1 CN 2022097635 W CN2022097635 W CN 2022097635W WO 2022257971 A1 WO2022257971 A1 WO 2022257971A1
Authority
WO
WIPO (PCT)
Prior art keywords
prediction residual
geometric
residual information
quantized
quantization
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2022/097635
Other languages
English (en)
French (fr)
Inventor
张伟
陈天
吕卓逸
杨付正
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Vivo Mobile Communication Co Ltd
Original Assignee
Vivo Mobile Communication Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Vivo Mobile Communication Co Ltd filed Critical Vivo Mobile Communication Co Ltd
Publication of WO2022257971A1 publication Critical patent/WO2022257971A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
    • H04N19/124—Quantisation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
    • H04N19/136—Incoming video signal characteristics or properties
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/90—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using coding techniques not provided for in groups H04N19/10-H04N19/85, e.g. fractals
    • H04N19/91—Entropy coding, e.g. variable length coding [VLC] or arithmetic coding

Definitions

  • the present application belongs to the technical field of point cloud processing, and in particular relates to a point cloud encoding and processing method, a point cloud decoding processing method and related equipment.
  • a point cloud is a form of representation of a three-dimensional object or scene, which is composed of a set of discrete point sets that are irregularly distributed in space and express the spatial structure and surface properties of a three-dimensional object or scene.
  • Point cloud data usually consists of geometric information describing a position such as three-dimensional coordinates (x, y, z) and attribute information of the position such as color (R, G, B) or reflectivity.
  • the encoding of geometric information and attribute information is carried out separately.
  • the geometric information of the point cloud will be preprocessed by rounding and deduplication, and the density of the source has a great influence on the preprocessing of the geometric information.
  • preprocessing the geometric information will drastically reduce the number of point clouds, resulting in the rate of the geometric code stream being greatly affected by the density of the source, resulting in a poor rate control effect of the geometric code stream of the point cloud.
  • the embodiment of the present application provides a point cloud encoding processing method, a point cloud decoding processing method and related equipment, which can solve the problem that the rate of the geometric code stream is greatly affected by the density of the source, which leads to the rate control of the geometric code stream of the point cloud. less effective problem.
  • a point cloud encoding processing method comprising:
  • Entropy coding is performed based on the quantized geometric prediction residual information to obtain a geometric code stream.
  • a point cloud decoding processing method comprising:
  • Predictive decoding is performed based on the geometric prediction residual information to obtain geometric information of the point cloud to be decoded.
  • a point cloud encoding processing device including:
  • the first encoding module is used to perform predictive encoding based on the geometric information of the point cloud to be encoded to obtain geometric prediction residual information;
  • a first quantization module configured to perform quantization processing on the geometric prediction residual information according to a geometric quantization parameter, to obtain quantized geometric prediction residual information
  • the second encoding module is configured to perform entropy encoding based on the quantized geometric prediction residual information to obtain a geometric code stream.
  • a point cloud decoding processing device including:
  • the first decoding module is used to perform entropy decoding on the geometric code stream to obtain quantized geometric prediction residual information
  • a first inverse quantization module configured to perform inverse quantization processing on the quantized geometric prediction residual information according to a geometric quantization parameter, to obtain geometric prediction residual information
  • the second decoding module is configured to perform predictive decoding based on the geometric prediction residual information to obtain geometric information of the point cloud to be decoded.
  • a terminal includes a processor, a memory, and a program or instruction stored in the memory and operable on the processor.
  • the program or instruction When the program or instruction is executed by the processor Realize the steps of the method described in the first aspect; or, realize the steps of the method described in the second aspect when the program or instruction is executed by the processor.
  • a terminal including a processor and a communication interface, wherein the processor or the communication interface is used for:
  • Entropy coding is performed based on the quantized geometric prediction residual information to obtain a geometric code stream.
  • a terminal including a processor and a communication interface, wherein the processor or the communication interface is used for:
  • Predictive decoding is performed based on the geometric prediction residual information to obtain geometric information of the point cloud to be decoded.
  • a readable storage medium on which a program or an instruction is stored, and when the program or instruction is executed by a processor, the steps of the point cloud encoding processing method as described in the first aspect are implemented Or, when the program or instruction is executed by the processor, the steps of the point cloud decoding processing method as described in the second aspect are implemented.
  • a ninth aspect provides a chip, the chip includes a processor and a communication interface, the communication interface is coupled to the processor, and the processor is used to run programs or instructions to achieve the points described in the first aspect The steps of the cloud encoding processing method, or the steps of implementing the point cloud decoding processing method as described in the second aspect.
  • a computer program/program product is provided, the computer program/program product is stored in a non-volatile storage medium, and the program/program product is executed by at least one processor to implement the first aspect
  • a communication device configured to execute the steps of the point cloud encoding processing method as described in the first aspect, or to execute the point cloud decoding processing method as described in any one of the second aspect step.
  • predictive coding is performed based on the geometric information of the point cloud to be encoded to obtain geometric prediction residual information; the geometric prediction residual information is quantized according to the geometric quantization parameter to obtain quantized geometric prediction residual information; based on Entropy encoding is performed on the quantized geometric prediction residual information to obtain a geometric code stream.
  • the geometric prediction residual information is quantized by the geometric quantization parameter, which reduces the influence of the source density on the geometric code stream rate, and can improve the rate control effect of the geometric code stream of the point cloud.
  • Fig. 1 is one of framework schematic diagrams of a kind of point cloud AVS encoder
  • Fig. 2 is one of frame schematic diagrams of a kind of point cloud AVS decoder
  • FIG. 3 is a flow chart of a point cloud encoding processing method provided by an embodiment of the present application.
  • Fig. 4 is a second schematic diagram of a point cloud AVS encoder framework
  • Fig. 5 is a second schematic diagram of a point cloud AVS decoder framework
  • FIG. 6 is a flow chart of a point cloud decoding processing method provided by an embodiment of the present application.
  • Fig. 7 is one of the structural diagrams of a point cloud encoding processing device provided by the embodiment of the present application.
  • Fig. 8 is the second structural diagram of a point cloud encoding processing device provided by the embodiment of the present application.
  • Fig. 9 is the third structural diagram of a point cloud encoding processing device provided by the embodiment of the present application.
  • Fig. 10 is the fourth structural diagram of a point cloud encoding processing device provided by the embodiment of the present application.
  • Fig. 11 is one of the structural diagrams of a point cloud decoding processing device provided by the embodiment of the present application.
  • Fig. 12 is the second structural diagram of a point cloud decoding processing device provided by the embodiment of the present application.
  • FIG. 13 is a structural diagram of a communication device provided by an embodiment of the present application.
  • FIG. 14 is a structural diagram of a terminal provided by an embodiment of the present application.
  • first, second and the like in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific sequence or sequence. It is to be understood that the terms so used are interchangeable under appropriate circumstances such that the embodiments of the application are capable of operation in sequences other than those illustrated or described herein and that "first" and “second” distinguish objects. It is usually one category, and the number of objects is not limited. For example, there may be one or more first objects.
  • “and/or” in the description and claims indicates at least one of the connected objects, and the character “/" generally indicates that the related objects before and after are an "or” relationship.
  • the codec terminal corresponding to the codec method in the embodiment of the present application can be a terminal, and the terminal can also be called a terminal device or a user terminal (User Equipment, UE), and the terminal can be a mobile phone, a tablet computer (Tablet Personal Computer), a laptop Laptop Computer (Laptop Computer) or Notebook Computer, Personal Digital Assistant (Personal Digital Assistant, PDA), PDA, Netbook, Ultra-mobile Personal Computer (UMPC), Mobile Internet Device (Mobile Internet Device) , MID), augmented reality (augmented reality, AR)/virtual reality (virtual reality, VR) equipment, robots, wearable devices (Wearable Device) or vehicle-mounted equipment (VUE), pedestrian terminals (PUE) and other terminal-side devices, Wearable devices include: smart watches, bracelets, earphones, glasses, etc. It should be noted that, the embodiment of the present application does not limit the specific type of the terminal.
  • the geometric information and attribute information of point cloud are encoded separately.
  • coordinate transformation is performed on the geometric information so that all point clouds are contained in a bounding box, and then the coordinates are quantized.
  • Quantization mainly plays the role of scaling. Since quantization will round the geometric coordinates, the geometric information of some points will be the same, which is called duplicate points. It is determined whether to remove duplicate points according to the parameters. Quantization and removal of duplicate points are two steps. Also known as the voxelization process.
  • the bounding box is divided into 8 sub-cubes, and the non-empty sub-cubes continue to be divided until the unit cube with leaf nodes of 1x1x1 is obtained.
  • the number of points in the node is encoded to generate a binary code stream.
  • the current AVS geometry division sequence includes two types:
  • Depth-first traversal order When dividing the geometry into an octree, the first node of the current layer will be divided continuously until the leaf node obtained by division is a 1x1x1 unit cube and stop dividing the current node. According to this sequence, the subsequent nodes of the current layer are divided until the division of the nodes on the current layer is completed.
  • Attribute coding is mainly aimed at color and reflectance information. First, judge whether to perform color space conversion according to the parameters. If color space conversion is performed, the color information is converted from Red Green Blue (RGB) color space to brightness color (YUV) color space. Then, the geometrically reconstructed point cloud is recolored with the original point cloud so that the unencoded attribute information corresponds to the reconstructed geometric information.
  • RGB Red Green Blue
  • YUV brightness color
  • the nearest neighbor of the point to be predicted is searched using the geometric spatial relationship, and the reconstructed attribute value of the found neighbor is used to predict the point to be predicted to obtain the predicted attribute value, and then the The real attribute value and the predicted attribute value are differentiated to obtain the prediction residual, and finally the prediction residual is quantized and encoded to generate a binary code stream.
  • the AVS decoding process corresponds to the encoding process.
  • the AVS decoder framework is shown in FIG. 2 .
  • the AVS encoding framework can include two stages of preprocessing and encoding.
  • the processing of point clouds in the preprocessing stage can be called out-of-loop processing, and after the preprocessing is completed, that is, the processing of point clouds in the encoding stage can be called in-loop processing.
  • FIG. 3 is a flow chart of a point cloud encoding processing method provided in an embodiment of the present application. As shown in FIG. 3, the point cloud encoding processing method includes the following steps:
  • Step 101 Perform predictive coding based on the geometric information of the point cloud to be coded to obtain geometric prediction residual information.
  • the geometric information may include a geometric position.
  • a prediction candidate list can be established for the geometric information of the point cloud to be encoded, the best geometric prediction value can be selected from the prediction candidate list, and the best geometric prediction value and geometric information can be subtracted to obtain the geometric prediction residual information .
  • Each geometric prediction value in the prediction candidate list may correspond to a geometric prediction mode.
  • a prediction candidate list may be established in advance, and the prediction candidate list may include N geometric prediction values, where the N geometric prediction values correspond to the N geometric prediction modes one-to-one, and N is a positive integer greater than 1.
  • the prediction candidate list includes 4 geometric prediction values
  • the point cloud to be encoded is the fifth point cloud to be encoded in all point clouds
  • the point cloud located before the point cloud to be encoded can be used
  • the geometric information of the four point clouds to be encoded whose encoding sequence is 1 to 4 is used to determine the geometric prediction value.
  • the determination rule of the geometric prediction value can be that the first geometric prediction value is the sum of the geometric information of the 4 point clouds to be encoded; the second geometric prediction value is the minimum geometric information of the 4 point clouds to be encoded; the third The first geometric prediction value is the average of the geometric information of the four point clouds to be encoded; the fourth geometric prediction value is the difference between the geometric information of the fourth point cloud to be encoded and the geometric information of the third point cloud to be encoded.
  • the geometric information of the point cloud to be encoded can be characterized as the three-dimensional coordinates (x, y, z) of the point cloud to be encoded.
  • Step 102 Perform quantization processing on the geometric prediction residual information according to the geometric quantization parameter to obtain quantized geometric prediction residual information.
  • the geometric quantization control parameter indicates that quantization processing is enabled; in the case that the geometric quantization control parameter indicates that quantization processing is enabled, the geometric prediction residual information is quantized according to the geometric quantization parameter to obtain a quantized geometric prediction residual difference information, performing entropy encoding based on the quantized geometric prediction residual information to obtain a geometric code stream; when the geometric quantization control parameter indicates that quantization processing is not enabled, performing entropy encoding according to the geometric prediction residual information to obtain Geometry stream.
  • the geometric quantization parameter can be read from the configuration file, for example, the value of the parameter GeomQP in the configuration file can be read as the geometric quantization parameter.
  • Step 103 Perform entropy coding based on the quantized geometric prediction residual information to obtain a geometric code stream.
  • entropy encoding may be performed on the quantized geometric prediction residual information and the geometric prediction mode to obtain a geometric code stream.
  • the lossy quantization in this preprocessing will have the following problems: When the source distribution is relatively sparse and uniform, the number of quantized points can be uniformly reduced by setting different quantization steps, but when the source distribution is relatively dense and concentrated , a smaller quantization step size will lead to a sharp decrease in the number of points; lossy quantization is done outside the ring, and realized through the quantization step size, it can be considered as a downsampling operation on the original point cloud data, and the number of downsampled points is affected by Influenced by the information source, the quantized geometric code stream will also be affected by the information source.
  • the rate control of the geometric code stream cannot be achieved;
  • the preprocessed point cloud is greatly affected by the density of the source.
  • the quality of the geometric information of the preprocessed point cloud cannot be accurately controlled by adjusting the quantization step outside the loop.
  • the geometric prediction residual information is quantized by the geometric quantization parameter, and the lossy quantization in the loop is introduced in the point cloud coding, which reduces the influence of the density of the information source on the geometric code stream rate, and can Improve the rate control effect of the geometric code stream of the point cloud.
  • predictive coding is performed based on the geometric information of the point cloud to be encoded to obtain geometric prediction residual information; the geometric prediction residual information is quantized according to the geometric quantization parameter to obtain quantized geometric prediction residual information; based on Entropy encoding is performed on the quantized geometric prediction residual information to obtain a geometric code stream.
  • the geometric prediction residual information is quantized by the geometric quantization parameter, which reduces the influence of the degree of information source density on the geometric code stream rate, and can improve the rate control effect of the geometric code stream of the point cloud.
  • performing quantization processing on the geometric prediction residual information according to a geometric quantization parameter includes:
  • the geometric prediction residual information is quantized according to the geometric quantization parameter.
  • the geometric quantization control parameter indicates enabling quantization processing, which may be regarded as indicating enabling in-loop lossy quantization for geometric prediction residual information.
  • the geometry quantization control parameter can be read from the configuration file, for example, the value of the parameter geometry_enable_quantized_flag in the configuration file can be read as the geometry quantization control parameter.
  • the parameter geometry_enable_quantized_flag can be a newly introduced parameter in the gps (geometry parameters set) high-level syntax element. When the geometry quantization control parameter is configured as 1, it may indicate that quantization processing is enabled; when the geometry quantization control parameter is configured as 0, it may indicate that quantization processing is not enabled.
  • the geometric quantization control parameter determines whether to perform quantization processing on the geometric prediction residual information according to the geometric quantization parameter, which can improve the flexibility of geometric information encoding of the point cloud.
  • the method further includes:
  • the geometric quantization control parameter indicates that the quantization process is not enabled, which can be regarded as indicating that the in-loop lossy quantization is not enabled for the geometric prediction residual information.
  • entropy coding is performed according to the geometric prediction residual information, so that it can be determined according to the geometric quantization control parameter during the predictive coding process of the geometric information Whether to quantize the geometric prediction residual information can improve the flexibility of the geometric information encoding of the point cloud.
  • the predictive encoding of the geometric information of the point cloud to be encoded includes:
  • the point cloud to be encoded is divided into a first sub-point cloud to be encoded and a second sub-point cloud to be encoded;
  • the geometric encoding control parameter indicates the first encoding mode
  • the point cloud to be encoded can be divided into a first sub-point cloud to be encoded and a second sub-point cloud to be encoded according to the relationship between the node identifier corresponding to the point cloud to be encoded and a preset threshold.
  • the first sub-point cloud to be encoded may be a low-bit point cloud to be encoded
  • the second sub-point cloud to be encoded may be a high-bit point cloud to be encoded.
  • the geometric information of the high-bit point cloud to be encoded may include the high-bit coordinates of the octree
  • the geometric information of the low-bit point cloud to be encoded may include the low-bit coordinates of the octree.
  • octree construction is performed to realize octree coding
  • geometric prediction and residual quantization are performed to realize predictive coding.
  • the high-bit point cloud to be encoded is obtained through octree reconstruction
  • the low-bit point cloud to be encoded is obtained through inverse quantization and geometric reconstruction.
  • the node identification corresponding to the point cloud to be coded can be the code layer in the process of octree coding layers.
  • the preset threshold can be 5
  • all point clouds to be encoded can include 10 encoding layers
  • the point clouds to be encoded corresponding to the first encoding layer to the fourth encoding layer can be used as high-bit encoding point clouds
  • the point clouds to be coded corresponding to the 5th coding layer to the 10th coded layer are regarded as low-bit point clouds to be coded.
  • the preset threshold may be the value of the parameter octree_division_end_nodeSizeLog2[3].
  • the value of the parameter octree_division_end_nodeSizeLog2[3] can be read from the configuration file as the preset threshold.
  • the geometric quantization parameter is greater than or equal to the preset threshold, the geometric prediction residual information is all quantized to 0, and it may not be necessary to perform entropy coding on the quantized geometric prediction residual information.
  • the preset threshold matches the first quantization parameter of the out-of-loop quantization in the preprocessing, the lossy quantization of the point cloud is consistent with the existing quantization.
  • the geometric quantization parameter is smaller than the preset threshold, the geometric prediction residual information will not be quantized to 0, and entropy coding may be performed based on the quantized geometric prediction residual information.
  • the point cloud to be encoded for predictive encoding of geometric information is different, and the user can modify the encoding mode by setting the geometric encoding control parameter, thereby modifying the point cloud to be encoded
  • the encoding method can improve the flexibility of point cloud encoding.
  • performing entropy coding based on the quantized geometric prediction residual information includes:
  • Entropy coding is performed based on the target quantized geometric prediction residual information.
  • the at least two candidate geometric prediction residual information may include candidate geometric prediction residual information related to the quantized geometric prediction residual information, and candidate geometric prediction residual information not related to the quantized geometric prediction residual information residual information.
  • the at least two candidate geometric prediction residual information may include quantized geometric prediction residual information and a fixed value ⁇ 0, 0, 0 ⁇ .
  • a rate-distortion optimization algorithm is introduced to process the target quantized geometric prediction residual information, and entropy coding is performed based on the target quantized geometric prediction residual information, which can improve the lossy coding of geometric information. s efficiency.
  • the target quantized geometric prediction residual information is the candidate geometric prediction residual information with the smallest rate-distortion cost among the at least two candidate geometric prediction residual information.
  • the at least two candidate geometric prediction residual information can be stored in the form of a candidate list, and the first candidate geometric prediction residual information in the candidate list is used as the best candidate geometric prediction residual information; traversing the candidate geometric prediction residual information in the candidate list Candidate geometric prediction residual information; if the rate-distortion cost corresponding to the current candidate geometric prediction residual information is less than the rate-distortion cost corresponding to the best candidate geometric prediction residual information, update the current candidate geometric prediction residual information to the best candidate geometry The prediction residual information, otherwise, the best candidate geometric prediction residual information is not updated; after traversing the candidate list, the best candidate geometric prediction residual information is determined as the target quantized geometric prediction residual information. After the target quantized geometric prediction residual information is determined, the target quantized geometric prediction residual information may be input into an encoder for entropy encoding.
  • the candidate geometric prediction residual information with the smallest rate-distortion cost among the at least two candidate geometric prediction residual information is determined as the target quantized geometric prediction residual information, so that the lossy coding process of the geometric information can be optimized, Improve point cloud encoding efficiency.
  • the rate-distortion cost corresponding to the candidate geometric prediction residual information is determined based on the geometric distortion value and the first prediction residual code rate, and the geometric distortion value is used to characterize the geometry corresponding to the candidate geometric prediction residual information. Distortion, the first prediction residual code rate is used to represent the expected bit value for encoding the candidate geometric prediction residual information.
  • the rate-distortion cost corresponding to the candidate geometric prediction residual information may be positively correlated with both the geometric distortion value and the code rate of the first prediction residual.
  • the rate-distortion cost cost1 corresponding to the candidate geometric prediction residual information may be:
  • ⁇ 1 can represent the weight parameter of the code rate and distortion in the rate-distortion cost.
  • ⁇ 1 can be set to 0.4, 0.5 or 0.6, etc.
  • rate1 can represent the first predicted residual code rate
  • dist can represent the geometric distortion value .
  • the calculation formula of the geometric distortion value dist1 may be as follows:
  • the function normal1 means to obtain a norm of the expression
  • recPos means the geometric coordinate reconstructed by using the candidate geometric prediction residual information and the geometric prediction value
  • oriPos means the original geometric coordinate
  • the rate-distortion cost corresponding to the candidate geometric prediction residual information is determined based on the geometric distortion value and the first prediction residual code rate, and the rate-distortion cost corresponding to the candidate geometric prediction residual information can be determined more accurately.
  • the at least two candidate geometric prediction residual information include candidate geometric prediction residual information related to the quantized geometric prediction residual information, and candidate geometric prediction residual information not related to the quantized geometric prediction residual information.
  • Prediction residual information Prediction residual information
  • the entropy coding based on the target quantized geometric prediction residual information includes:
  • target quantized geometric prediction residual information is candidate geometric prediction residual information related to the quantized geometric prediction residual information, based on the identifier corresponding to the target quantized geometric prediction residual information and the target quantization Entropy encoding of geometric prediction residual information;
  • target quantized geometric prediction residual information is candidate geometric prediction residual information unrelated to the quantized geometric prediction residual information
  • entropy encoding is performed based on an identifier corresponding to the target quantized geometric prediction residual information.
  • each candidate geometric prediction residual information may be correspondingly provided with an identifier.
  • the candidate geometric prediction residual information is related to the quantized geometric prediction residual information. It may be that the candidate geometric prediction residual information may be obtained based on the quantized geometric prediction residual information. For example, the candidate geometric prediction residual information is equal to the quantized geometric prediction residual information.
  • the identifier corresponding to the candidate geometric prediction residual information may be 1; or, the candidate geometric prediction residual information is equal to an integer multiple of the quantized geometric prediction residual information, and the corresponding identifier for the candidate geometric prediction residual information may be 2, etc.; the candidate geometric prediction residual information is not related to the quantized geometric prediction residual information, it may be that the candidate geometric prediction residual information is preset geometric prediction residual information, for example, it may be (0, 0 , 0), the identifier corresponding to the candidate geometric prediction residual information may be 0.
  • a geometric rate-distortion optimization control parameter may be set, and if the geometric rate-distortion optimization control parameter is a first preset value, performing entropy encoding based on the target quantized geometric prediction residual information includes: If the geometric prediction residual information is candidate geometric prediction residual information related to the quantized geometric prediction residual information, based on the identifier corresponding to the target quantized geometric prediction residual information and the target quantized geometric prediction residual information Perform entropy encoding; in the case that the target quantized geometric prediction residual information is candidate geometric prediction residual information unrelated to the quantized geometric prediction residual information, based on the identification corresponding to the target quantized geometric prediction residual information Do entropy encoding.
  • performing entropy encoding based on the target quantized geometric prediction residual information includes: when the target quantized geometric prediction residual information is the same as the quantized geometric prediction In the case of the candidate geometric prediction residual information related to the residual information, entropy encoding is performed based on the target quantized geometric prediction residual information; when the target quantized geometric prediction residual information is different from the quantized geometric prediction residual information In the case of relevant candidate geometric prediction residual information, entropy encoding is performed based on the target quantized geometric prediction residual information.
  • This embodiment does not limit the first preset value and the second preset value.
  • the first preset value may be 1, and the second preset value may be 0.
  • the target quantized geometric prediction residual information may be determined whether the target quantized geometric prediction residual information is related or not related to the quantized geometric prediction residual information through an identifier corresponding to the target quantized geometric prediction residual information.
  • the identifier corresponding to the target quantized geometric prediction residual information may be analyzed first, and if the target quantized geometric prediction residual information and the quantized geometric prediction residual If the difference information is irrelevant, the target quantized geometric prediction residual information can be found according to the identifier corresponding to the target quantized geometric prediction residual information; if the target quantized geometric prediction residual information is determined according to the identifier corresponding to the target quantized geometric prediction residual information
  • the target quantized geometric prediction residual information can be obtained by decoding from the geometric code stream.
  • the target quantized geometric prediction residual information is candidate geometric prediction residual information related to the quantized geometric prediction residual information, based on the identifier corresponding to the target quantized geometric prediction residual information performing entropy encoding with the target quantized geometric prediction residual information; in the case that the target quantized geometric prediction residual information is candidate geometric prediction residual information unrelated to the quantized geometric prediction residual information, based on the Entropy encoding is performed on the identifier corresponding to the target quantized geometric prediction residual information.
  • part of the target quantized geometric prediction residual information may not be encoded, but only the identifier corresponding to the target quantized geometric prediction residual information may be encoded, which can further improve the encoding efficiency.
  • the predictive encoding of the geometric information of the point cloud to be encoded includes:
  • Predictive encoding is performed on the geometric information of the point cloud to be encoded corresponding to the quantized point cloud obtained after deduplication processing.
  • coordinate translation processing can be performed on the point cloud to be encoded. Coordinate origin (0, 0, 0), where the bounding box represents the smallest cuboid that contains all points in the input point cloud.
  • the first quantization step size can be preset by the user.
  • the first quantization step size QS can be:
  • i 0
  • (X, Y, Z) represents the quantized coordinates of the quantized point cloud
  • (x, y, z) represents the coordinates of the original point cloud
  • QS represents the first quantization step size
  • the round(s) function represents returning the nearest Integer
  • the quantized coordinates of the quantized point cloud are calculated, there may be cases where the quantized coordinates of multiple original point clouds are the same, and the original point clouds with the same quantized coordinates are repeated points.
  • deduplication processing can be performed on the quantized point cloud.
  • the point cloud coordinates to be encoded corresponding to the quantized point cloud obtained after deduplication processing are the original point cloud coordinates, and no quantization operation is performed on the point cloud coordinates.
  • the geometric information of the point cloud to be encoded corresponding to the quantized point cloud obtained after deduplication processing is predicted and encoded, so that only the number of point clouds is down-sampled during out-of-loop quantization, and the point cloud coordinates are not sampled. Quantify.
  • performing quantization processing on the geometric prediction residual information according to a geometric quantization parameter includes:
  • the first geometric quantization step size QS1 can be:
  • 2 shift1 may represent the first preset geometric offset value
  • shift1 may represent the number of bits shifted during the quantization process
  • QP1 may represent the geometric quantization parameter
  • shift1 can be configured as 14.
  • the quantized geometric prediction residual information QtRes1 obtained by quantization processing can be:
  • Res1 may represent geometric prediction residual information
  • offset1 may represent half of the first preset geometric offset value, that is, offset1 is 2 shift1-1 , and a rounding operation may be realized through offset1.
  • the first geometric quantization step is determined according to the geometric quantization parameter, and the geometric prediction residual information is quantized based on the first geometric quantization step and the first preset geometric offset value, so that a relatively Good quantization effect.
  • the geometric prediction residual information includes three-dimensional sub-geometric prediction residual information
  • the geometric quantization parameters include three sub-geometric quantization parameters respectively corresponding to the sub-geometric prediction residual information of the three dimensions.
  • the three dimensions may be respectively X, Y, and Z dimensions in the three-dimensional coordinate system.
  • Three sub-geometry quantization parameters can be configured through the parameter GeomQP[3] in the configuration file (cfg), and the three sub-geometry quantization parameters can perform corresponding quantization on the three-dimensional sub-geometry prediction residual information.
  • the geometric quantization parameters include three sub-geometric quantization parameters respectively corresponding to the three-dimensional sub-geometric prediction residual information, which can respectively quantize the three-dimensional sub-geometric prediction residual information, so that The robustness and adaptability of lossy quantization within the geometric information loop can be improved.
  • the method also includes:
  • Predictive encoding is performed on the attribute information of the point cloud to be encoded to obtain attribute prediction residual information
  • Entropy coding is performed based on the quantized attribute prediction residual information to obtain an attribute code stream.
  • the first attribute quantization step size may be determined according to the attribute quantization parameter; and the attribute prediction residual information is quantized based on the first attribute quantization step size and a preset attribute offset value.
  • the quantization step size QS2 of the first attribute may be:
  • QP2 may represent an attribute quantization parameter.
  • Quantized attribute prediction residual information QtRes2 obtained by quantization processing can be:
  • Res2 may represent attribute prediction residual information
  • offset2 may represent a preset attribute offset value, for example, offset2 may be set to 0.5.
  • a prediction candidate list can be established for the attribute information of the point cloud to be encoded, and the best attribute prediction value can be selected from the prediction candidate list, and the best attribute prediction value and attribute information can be subtracted to obtain the attribute prediction residual. poor information.
  • Each attribute prediction value in the prediction candidate list may correspond to an attribute prediction mode.
  • a prediction candidate list may be established in advance, and the prediction candidate list may include N attribute prediction values, where N attribute prediction values correspond to N attribute prediction modes one-to-one, and N is a positive integer greater than 1.
  • the prediction candidate list includes 4 attribute prediction values
  • the point cloud to be encoded is the fifth point cloud to be encoded in all point clouds
  • the point cloud located before the point cloud to be encoded can be used
  • the attribute information of the four point clouds to be encoded whose encoding sequence is 1 to 4 is used to determine the attribute prediction value.
  • the determination rule of the attribute prediction value can be that the first attribute prediction value is the sum of the attribute information of the 4 point clouds to be encoded; the second attribute prediction value is the minimum attribute information of the 4 point clouds to be encoded; the third The predicted value of the first attribute is the average value of the attribute information of the four point clouds to be encoded; the predicted value of the fourth attribute is the difference between the attribute information of the fourth point cloud to be encoded and the attribute information of the third point cloud to be encoded.
  • the attribute information of the point cloud to be encoded can be represented as the three-dimensional coordinates (x, y, z) of the point cloud to be encoded.
  • predictive encoding is performed on the attribute information of the point cloud to be encoded to obtain attribute prediction residual information; the attribute prediction residual information is quantized according to attribute quantization parameters to obtain quantized attribute prediction residual information; Entropy encoding is performed based on the quantized attribute prediction residual information to obtain an attribute code stream.
  • performing entropy coding based on the quantized attribute prediction residual information includes:
  • Entropy coding is performed based on the target quantization attribute prediction residual information.
  • the at least two candidate attribute prediction residual information may include candidate attribute prediction residual information related to the quantized attribute prediction residual information, and candidate attribute prediction residual information not related to the quantized attribute prediction residual information residual information.
  • the at least two candidate attribute prediction residual information may include quantization attribute prediction residual information and a fixed value ⁇ 0, 0, 0 ⁇ .
  • a rate-distortion optimization algorithm is introduced to process the target quantization attribute prediction residual information, and entropy coding is performed based on the target quantization attribute prediction residual information, which can improve the lossy coding of attribute information. s efficiency.
  • the target quantized attribute prediction residual information is the candidate attribute prediction residual information with the smallest rate-distortion cost among the at least two candidate attribute prediction residual information.
  • the at least two candidate attribute prediction residual information may be stored in the form of a candidate list, and the first candidate attribute prediction residual information in the candidate list is used as the best candidate attribute prediction residual information;
  • Candidate attribute prediction residual information if the rate-distortion cost corresponding to the current candidate attribute prediction residual information is less than the rate-distortion cost corresponding to the best candidate attribute prediction residual information, update the current candidate attribute prediction residual information to the best candidate attribute
  • the prediction residual information otherwise, the best candidate attribute prediction residual information is not updated; after traversing the candidate list, the best candidate attribute prediction residual information is determined as the target quantized attribute prediction residual information.
  • the target quantization attribute prediction residual information may be input into an encoder for entropy encoding.
  • the candidate attribute prediction residual information with the smallest rate-distortion cost among the at least two candidate attribute prediction residual information is determined as the target quantized attribute prediction residual information, so that the lossy coding process of the attribute information can be optimized, Improve point cloud encoding efficiency.
  • the rate-distortion cost corresponding to the candidate attribute prediction residual information is determined based on the attribute distortion value and the second prediction residual code rate, and the attribute distortion value is used to characterize the attribute corresponding to the candidate attribute prediction residual information Distortion, the second prediction residual code rate is used to represent the expected bit value for encoding the candidate attribute prediction residual information.
  • the rate-distortion cost corresponding to the candidate attribute prediction residual information may be positively correlated with both the attribute distortion value and the second prediction residual code rate.
  • the rate-distortion cost cost2 corresponding to the candidate attribute prediction residual information may be:
  • ⁇ 2 can represent the weight parameter of the code rate and distortion in the rate-distortion cost.
  • ⁇ 2 can be set to 0.4, 0.5 or 0.6, etc.
  • rate2 can represent the second predicted residual code rate
  • dist2 can represent the attribute distortion value .
  • the formula for calculating the attribute distortion value dist2 can be as follows:
  • the function normal1 means to obtain a norm of the expression
  • recAttri means the reconstruction attribute value obtained by using the candidate attribute prediction residual information and the attribute prediction value
  • oriAttri means the original attribute value
  • the rate-distortion cost corresponding to the candidate attribute prediction residual information is determined based on the attribute distortion value and the second prediction residual code rate, and the rate-distortion cost corresponding to the candidate attribute prediction residual information can be determined more accurately.
  • the at least two candidate attribute prediction residual information include candidate attribute prediction residual information related to the quantized attribute prediction residual information, and candidate attributes not related to the quantized attribute prediction residual information Prediction residual information;
  • the entropy coding based on the target quantization attribute prediction residual information includes:
  • the target quantized attribute prediction residual information is candidate attribute prediction residual information related to the quantized attribute prediction residual information, based on the identifier corresponding to the target quantized geometric prediction residual information and the target quantization Attribute prediction residual information is entropy encoded;
  • target quantized property prediction residual information is candidate property prediction residual information unrelated to the quantized property prediction residual information
  • entropy coding is performed based on an identifier corresponding to the target quantized property prediction residual information.
  • the candidate attribute prediction residual information is related to the quantized attribute prediction residual information, and the candidate attribute prediction residual information may be obtained based on the quantized attribute prediction residual information.
  • the candidate attribute prediction residual information is equal to Quantized attribute prediction residual information, or candidate attribute prediction residual information is equal to an integer multiple of quantized attribute prediction residual information, etc.; candidate attribute prediction residual information is not related to the quantized attribute prediction residual information, which can be, candidate
  • the attribute prediction residual information is preset attribute prediction residual information, for example, may be (0, 0, 0).
  • an attribute rate-distortion optimization control parameter may be set, and if the attribute rate-distortion optimization control parameter is a third preset value, performing entropy coding on the residual information based on the target quantization attribute prediction includes: If the attribute prediction residual information is candidate attribute prediction residual information related to the quantized attribute prediction residual information, based on the identifier corresponding to the target quantized attribute prediction residual information and the target quantized attribute prediction residual information Perform entropy encoding; when the target quantized attribute prediction residual information is candidate attribute prediction residual information that is not related to the quantized attribute prediction residual information, based on the identifier corresponding to the target quantized attribute prediction residual information Do entropy encoding.
  • performing entropy encoding based on the target quantization property prediction residual information includes: when the target quantization property prediction residual information is the same as the quantization property prediction In the case of the candidate attribute prediction residual information related to the residual information, entropy coding is performed based on the target quantization attribute prediction residual information; when the target quantization attribute prediction residual information is different from the quantization attribute prediction residual information In the case of relevant candidate attribute prediction residual information, entropy coding is performed based on the target quantization attribute prediction residual information.
  • This embodiment does not limit the third preset value and the fourth preset value.
  • the third preset value may be 1, and the fourth preset value may be 0.
  • the target quantitative attribute prediction residual information is related or not related to the quantization attribute prediction residual information through an identifier corresponding to the target quantitative attribute prediction residual information.
  • the identifier corresponding to the target quantization attribute prediction residual information may be analyzed first, if the target quantization attribute prediction residual information and the quantization attribute prediction residual information are determined according to the identifier corresponding to the target quantization attribute prediction residual information If the difference information is irrelevant, the target quantization attribute prediction residual information can be found according to the identifier corresponding to the target quantization attribute prediction residual information; if the target quantization attribute prediction residual information is determined according to the identifier corresponding to the target quantization attribute prediction residual information Related to the quantization attribute prediction residual information, the target quantization attribute prediction residual information can be obtained by decoding the attribute code stream.
  • the target quantization attribute prediction residual information is candidate attribute prediction residual information related to the quantization attribute prediction residual information, based on the identifier corresponding to the target quantization attribute prediction residual information performing entropy encoding with the target quantization attribute prediction residual information; in the case that the target quantization attribute prediction residual information is candidate attribute prediction residual information unrelated to the quantization attribute prediction residual information, based on the Entropy encoding is performed on the identification corresponding to the target quantization attribute prediction residual information.
  • part of the target quantization attribute prediction residual information may not be encoded, but only the identifier corresponding to the target quantization attribute prediction residual information may be encoded, which can further improve encoding efficiency.
  • FIG. 6 is a flow chart of a point cloud decoding processing method provided in an embodiment of the present application. As shown in FIG. 6, the point cloud decoding processing method includes the following steps:
  • Step 201 performing entropy decoding on the geometric code stream to obtain quantized geometric prediction residual information
  • Step 202 Dequantize the quantized geometric prediction residual information according to the geometric quantization parameter to obtain geometric prediction residual information
  • Step 203 Perform predictive decoding based on the geometric prediction residual information to obtain geometric information of the point cloud to be decoded.
  • entropy decoding may be performed on the geometric code stream to obtain quantized geometric prediction residual information and a geometric prediction mode.
  • Predictive decoding may be performed based on the geometric prediction residual information and the geometric prediction mode to obtain geometric information of the point cloud to be decoded.
  • the geometric prediction mode can be analyzed, and the corresponding geometric prediction value can be selected according to the geometric prediction mode; the geometric prediction value and the residual information of the geometric prediction can be added to obtain the geometric information of the point cloud to be decoded.
  • Geometric information may include geometric coordinates.
  • performing entropy decoding on the geometric code stream to obtain quantized geometric prediction residual information includes:
  • the geometry quantization control parameter indicates that quantization processing is enabled, entropy decoding is performed on the geometry code stream to obtain quantized geometry prediction residual information.
  • the method further includes:
  • geometry quantization control parameter indicates that quantization processing is not enabled, entropy decoding is performed on the geometry code stream to obtain geometry prediction residual information.
  • performing dequantization processing on the quantized geometric prediction residual information according to geometric quantization parameters includes:
  • Inverse quantization processing is performed on the quantized geometric prediction residual information based on the second geometric quantization step size and a second preset geometric offset value.
  • the second geometric quantization step size QS3 can be:
  • 2 shift3 may represent the second preset geometric offset value
  • shift3 may represent the number of bits shifted during the quantization process, the larger the shift3 is, the more accurate the quantization result is
  • QP1 may represent the geometric quantization parameter
  • shift3 can be configured as 6.
  • the geometric prediction residual information RQtRes1 obtained by inverse quantization processing can be:
  • QtRes1 may represent quantized geometric prediction residual information
  • offset3 may represent half of the second preset geometric offset value, that is, offset3 is 2 shift3-1 .
  • the geometric prediction residual information includes three-dimensional sub-geometric prediction residual information
  • the geometric quantization parameters include three sub-geometric quantization parameters respectively corresponding to the sub-geometric prediction residual information of the three dimensions.
  • the method also includes:
  • Predictive decoding is performed based on the attribute prediction residual information to obtain attribute information of the point cloud to be decoded.
  • the entropy decoding the attribute code stream to obtain the quantized attribute prediction residual information may include: determining whether the attribute quantization control parameter indicates enabling quantization processing; in the case that the attribute quantization control parameter indicates enabling quantization processing, Entropy decoding is performed on the attribute code stream to obtain quantized attribute prediction residual information; when the attribute quantization control parameter indicates that quantization processing is not enabled, entropy decoding is performed on the attribute code stream to obtain attribute prediction residual information.
  • the attribute code stream can be entropy decoded to obtain the quantized attribute prediction residual information and the attribute prediction mode.
  • Predictive decoding may be performed based on the attribute prediction residual information and the attribute prediction mode to obtain attribute information of the point cloud to be decoded.
  • the attribute prediction mode can be analyzed, and the corresponding attribute prediction value can be selected according to the attribute prediction mode; the attribute prediction value and the attribute prediction residual information can be added to obtain the attribute information of the point cloud to be decoded.
  • the attribute information may include attribute coordinates.
  • the second attribute quantization step size can be determined according to the attribute quantization parameter, and the second attribute quantization step size QS4 can be:
  • QP2 may represent an attribute quantization parameter.
  • the attribute prediction residual information RQtRes2 obtained by inverse quantization processing can be:
  • QtRes2 may represent quantized attribute prediction residual information.
  • this embodiment is an implementation manner of the decoding side corresponding to the embodiment shown in FIG. The embodiment will not be repeated, and the same beneficial effect can also be achieved.
  • the point cloud encoding processing method provided in the embodiment of the present application may be executed by a point cloud encoding processing device, or a control module in the point cloud encoding processing device for executing the point cloud encoding processing method.
  • the point cloud coding processing device provided in the embodiment of the present application is described by taking the method for performing the point cloud coding processing by the point cloud coding processing device as an example.
  • FIG. 7 is one of the structural diagrams of a point cloud encoding processing device provided in the embodiment of the present application.
  • the point cloud encoding processing device 300 includes:
  • the first encoding module 301 is configured to perform predictive encoding based on the geometric information of the point cloud to be encoded to obtain geometric prediction residual information;
  • the first quantization module 302 is configured to perform quantization processing on the geometric prediction residual information according to a geometric quantization parameter to obtain quantized geometric prediction residual information;
  • the second encoding module 303 is configured to perform entropy encoding based on the quantized geometric prediction residual information to obtain a geometric code stream.
  • the first quantization module 302 is specifically configured to:
  • the geometric prediction residual information is quantized according to the geometric quantization parameter.
  • the first quantization module 302 is specifically further configured to:
  • the first encoding module 301 is specifically configured to:
  • the point cloud to be encoded is divided into a first sub-point cloud to be encoded and a second sub-point cloud to be encoded;
  • the geometric encoding control parameter indicates the first encoding mode
  • the second encoding module 303 specifically includes:
  • a first determining unit 3031 configured to determine at least two candidate geometric prediction residual information based on the quantized geometric prediction residual information
  • the first obtaining unit 3032 is configured to obtain the rate-distortion cost corresponding to the at least two candidate geometric prediction residual information
  • the second determining unit 3033 is configured to determine the target quantized geometric prediction residual information according to the rate-distortion cost corresponding to the at least two candidate geometric prediction residual information;
  • the first coding unit 3034 is configured to perform entropy coding based on the target quantized geometric prediction residual information.
  • the target quantized geometric prediction residual information is the candidate geometric prediction residual information with the smallest rate-distortion cost among the at least two candidate geometric prediction residual information.
  • the rate-distortion cost corresponding to the candidate geometric prediction residual information is determined based on the geometric distortion value and the first prediction residual code rate, and the geometric distortion value is used to characterize the geometry corresponding to the candidate geometric prediction residual information. Distortion, the first prediction residual code rate is used to represent the expected bit value for encoding the candidate geometric prediction residual information.
  • the at least two candidate geometric prediction residual information include candidate geometric prediction residual information related to the quantized geometric prediction residual information, and candidate geometric prediction residual information not related to the quantized geometric prediction residual information.
  • Prediction residual information Prediction residual information
  • the first coding unit 3034 is specifically used for:
  • target quantized geometric prediction residual information is candidate geometric prediction residual information related to the quantized geometric prediction residual information, based on the identifier corresponding to the target quantized geometric prediction residual information and the target quantization Entropy encoding of geometric prediction residual information;
  • target quantized geometric prediction residual information is candidate geometric prediction residual information unrelated to the quantized geometric prediction residual information
  • entropy encoding is performed based on an identifier corresponding to the target quantized geometric prediction residual information.
  • the first encoding module 301 is specifically configured to:
  • Predictive encoding is performed on the geometric information of the point cloud to be encoded corresponding to the quantized point cloud obtained after deduplication processing.
  • the first quantization module 302 is specifically configured to:
  • the geometric prediction residual information includes three-dimensional sub-geometric prediction residual information
  • the geometric quantization parameters include three sub-geometric quantization parameters respectively corresponding to the sub-geometric prediction residual information of the three dimensions.
  • the device 300 further includes:
  • the third encoding module 304 is configured to perform predictive encoding on the attribute information of the point cloud to be encoded to obtain attribute prediction residual information;
  • the second quantization module 305 is configured to perform quantization processing on the attribute prediction residual information according to the attribute quantization parameter, to obtain quantized attribute prediction residual information;
  • the fourth encoding module 306 is configured to perform entropy encoding based on the quantized attribute prediction residual information to obtain an attribute code stream.
  • the fourth encoding module 306 specifically includes:
  • the third determining unit 3061 is configured to determine at least two candidate attribute prediction residual information based on the quantized attribute prediction residual information
  • the second obtaining unit 3062 is configured to obtain the rate-distortion cost corresponding to the at least two candidate attribute prediction residual information
  • the fourth determining unit 3063 is configured to determine the target quantized attribute prediction residual information according to the rate-distortion cost corresponding to the at least two candidate attribute prediction residual information;
  • the second encoding unit 3064 is configured to perform entropy encoding based on the target quantization attribute prediction residual information.
  • the target quantized attribute prediction residual information is the candidate attribute prediction residual information with the smallest rate-distortion cost among the at least two candidate attribute prediction residual information.
  • the rate-distortion cost corresponding to the candidate attribute prediction residual information is determined based on the attribute distortion value and the second prediction residual code rate, and the attribute distortion value is used to characterize the attribute corresponding to the candidate attribute prediction residual information Distortion, the second prediction residual code rate is used to represent the expected bit value for encoding the candidate attribute prediction residual information.
  • the at least two candidate attribute prediction residual information include candidate attribute prediction residual information related to the quantized attribute prediction residual information, and candidate attributes not related to the quantized attribute prediction residual information Prediction residual information;
  • the second coding unit 3064 is specifically used for:
  • the target quantized attribute prediction residual information is candidate attribute prediction residual information related to the quantized attribute prediction residual information, based on the identifier corresponding to the target quantized geometric prediction residual information and the target quantization Attribute prediction residual information is entropy encoded;
  • target quantized property prediction residual information is candidate property prediction residual information unrelated to the quantized property prediction residual information
  • entropy coding is performed based on an identifier corresponding to the target quantized property prediction residual information.
  • the point cloud encoding processing device 300 in the embodiment of the present application can improve the rate control effect of the geometric code stream of the point cloud.
  • the point cloud encoding processing device in the embodiment of the present application may be a device, a device with an operating system or an electronic device, or a component, an integrated circuit, or a chip in a terminal.
  • the apparatus or electronic equipment may be a mobile terminal or a non-mobile terminal.
  • a mobile terminal may include but not limited to the types of terminals listed above, and a non-mobile terminal may be a server, a network attached storage (Network Attached Storage, NAS), a personal computer (personal computer, PC), a television (television , TV), teller machines or self-service machines, etc., are not specifically limited in this embodiment of the present application.
  • the point cloud encoding processing device provided by the embodiment of the present application can realize each process realized by the method embodiment in FIG. 3 and achieve the same technical effect. In order to avoid repetition, details are not repeated here.
  • the point cloud decoding processing method provided by the embodiment of the present application may be executed by a point cloud decoding processing device, or a control module in the point cloud decoding processing device for executing the point cloud decoding processing method.
  • the point cloud decoding processing device provided in the embodiment of the present application is described by taking the method for performing the point cloud decoding processing by the point cloud decoding processing device as an example.
  • FIG. 11 is one of the structural diagrams of a point cloud decoding processing device provided in the embodiment of the present application.
  • the point cloud decoding processing device 400 includes:
  • the first decoding module 401 is configured to perform entropy decoding on the geometric code stream to obtain quantized geometric prediction residual information
  • the first inverse quantization module 402 is configured to perform inverse quantization processing on the quantized geometric prediction residual information according to the geometric quantization parameter to obtain the geometric prediction residual information;
  • the second decoding module 403 is configured to perform predictive decoding based on the geometric prediction residual information to obtain geometric information of the point cloud to be decoded.
  • the first decoding module 401 is specifically configured to:
  • the geometry quantization control parameter indicates that quantization processing is enabled, entropy decoding is performed on the geometry code stream to obtain quantized geometry prediction residual information.
  • the first decoding module 401 is specifically further configured to:
  • geometry quantization control parameter indicates that quantization processing is not enabled, entropy decoding is performed on the geometry code stream to obtain geometry prediction residual information.
  • the first inverse quantization module 402 is specifically configured to:
  • Inverse quantization processing is performed on the quantized geometric prediction residual information based on the second geometric quantization step size and the second preset geometric offset value.
  • the geometric prediction residual information includes three-dimensional sub-geometric prediction residual information
  • the geometric quantization parameters include three sub-geometric quantization parameters respectively corresponding to the sub-geometric prediction residual information of the three dimensions.
  • the device 400 further includes:
  • the third decoding module 404 is configured to perform entropy decoding on the attribute code stream to obtain quantized attribute prediction residual information
  • the second inverse quantization module 405 is configured to perform inverse quantization processing on the quantized attribute prediction residual information according to the attribute quantization parameter, to obtain attribute prediction residual information;
  • the fourth decoding module 406 is configured to perform predictive decoding based on the attribute prediction residual information to obtain attribute information of the point cloud to be decoded.
  • the point cloud decoding processing device 400 in the embodiment of the present application can improve the rate control effect of the geometric code stream of the point cloud.
  • the point cloud decoding processing device in the embodiment of the present application may be a device, a device with an operating system or an electronic device, or a component, an integrated circuit, or a chip in a terminal.
  • the apparatus or electronic equipment may be a mobile terminal or a non-mobile terminal.
  • a mobile terminal may include but not limited to the types of terminals listed above, and a non-mobile terminal may be a server, a network attached storage (Network Attached Storage, NAS), a personal computer (personal computer, PC), a television (television , TV), teller machines or self-service machines, etc., are not specifically limited in this embodiment of the present application.
  • the point cloud decoding processing device provided in the embodiment of the present application can realize each process realized by the method embodiment in FIG. 6 and achieve the same technical effect. In order to avoid repetition, details are not repeated here.
  • this embodiment of the present application further provides a communication device 500, including a processor 501, a memory 502, and programs or instructions stored in the memory 502 and operable on the processor 501,
  • a communication device 500 including a processor 501, a memory 502, and programs or instructions stored in the memory 502 and operable on the processor 501
  • the communication device 500 is a terminal
  • the program or instruction is executed by the processor 501
  • each process of the above-mentioned point cloud encoding processing method embodiment can be realized, and the same technical effect can be achieved; or, the program or instruction is executed by the processor 501
  • the processor 501 When 501 is executed, each process of the above-mentioned point cloud decoding processing method embodiment is realized, and the same technical effect can be achieved. To avoid repetition, details are not repeated here.
  • the embodiment of the present application also provides a terminal, including a processor and a communication interface.
  • the embodiment of the terminal corresponds to the above-mentioned embodiment of the point cloud encoding processing method, or, the embodiment of the terminal corresponds to the embodiment of the above-mentioned point cloud decoding processing method.
  • each implementation process and implementation manner of the foregoing method embodiments can be applied to this terminal embodiment, and can achieve the same technical effect.
  • FIG. 14 is a schematic diagram of a hardware structure of a terminal implementing an embodiment of the present application.
  • the terminal 600 includes but is not limited to: a radio frequency unit 601, a network module 602, an audio output unit 603, an input unit 604, a sensor 605, a display unit 606, a user input unit 607, an interface unit 608, a memory 609, and a processor 610, etc. at least some of the components.
  • the terminal 600 can also include a power supply (such as a battery) for supplying power to various components, and the power supply can be logically connected to the processor 610 through the power management system, so as to manage charging, discharging, and power consumption management through the power management system and other functions.
  • a power supply such as a battery
  • the terminal structure shown in FIG. 6 does not constitute a limitation on the terminal.
  • the terminal may include more or fewer components than shown in the figure, or combine some components, or arrange different components, which will not be repeated here.
  • the input unit 604 may include a graphics processor (Graphics Processing Unit, GPU) 6041 and a microphone 6042, and the graphics processor 6041 is used for the image capture device (such as the image data of the still picture or video obtained by the camera) for processing.
  • the display unit 606 may include a display panel 6061, and the display panel 6061 may be configured in the form of a liquid crystal display, an organic light emitting diode, or the like.
  • the user input unit 607 includes a touch panel 6071 and other input devices 6072 .
  • the touch panel 6071 is also called a touch screen.
  • the touch panel 6071 may include two parts, a touch detection device and a touch controller.
  • Other input devices 6072 may include, but are not limited to, physical keyboards, function keys (such as volume control buttons, switch buttons, etc.), trackballs, mice, and joysticks, which will not be repeated here.
  • the radio frequency unit 601 receives the downlink data from the network side device, and processes it to the processor 610; in addition, sends the uplink data to the network side device.
  • the radio frequency unit 601 includes, but is not limited to, an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier, a duplexer, and the like.
  • the memory 609 can be used to store software programs or instructions as well as various data.
  • the memory 609 may mainly include a program or instruction storage area and a data storage area, wherein the program or instruction storage area may store an operating system, an application program or instructions required by at least one function (such as a sound playback function, an image playback function, etc.) and the like.
  • the memory 609 may include a high-speed random access memory, and may also include a nonvolatile memory, wherein the nonvolatile memory may be a read-only memory (Read-Only Memory, ROM), a programmable read-only memory (Programmable ROM) , PROM), erasable programmable read-only memory (Erasable PROM, EPROM), electrically erasable programmable read-only memory (Electrically EPROM, EEPROM) or flash memory.
  • ROM Read-Only Memory
  • PROM programmable read-only memory
  • PROM erasable programmable read-only memory
  • Erasable PROM Erasable PROM
  • EPROM electrically erasable programmable read-only memory
  • EEPROM electrically erasable programmable read-only memory
  • flash memory for example at least one magnetic disk storage device, flash memory device, or other non-volatile solid-state storage device.
  • the processor 610 may include one or more processing units; optionally, the processor 610 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface and application programs or instructions, etc., Modem processors mainly handle wireless communications, such as baseband processors. It can be understood that the foregoing modem processor may not be integrated into the processor 610 .
  • the terminal is used to execute the point cloud encoding processing method:
  • the processor or the communication interface is used to: perform predictive encoding based on the geometric information of the point cloud to be encoded to obtain geometric prediction residual information; perform quantization processing on the geometric prediction residual information according to geometric quantization parameters to obtain quantized geometry Prediction residual information; performing entropy coding based on the quantized geometric prediction residual information to obtain a geometric code stream.
  • processor 610 is also used for:
  • the geometric prediction residual information is quantized according to the geometric quantization parameter.
  • processor 610 is also used for:
  • processor 610 is also used for:
  • the point cloud to be encoded is divided into a first sub-point cloud to be encoded and a second sub-point cloud to be encoded;
  • the geometric encoding control parameter indicates the first encoding mode
  • processor 610 is also used for:
  • Entropy coding is performed based on the target quantized geometric prediction residual information.
  • the target quantized geometric prediction residual information is the candidate geometric prediction residual information with the smallest rate-distortion cost among the at least two candidate geometric prediction residual information.
  • the rate-distortion cost corresponding to the candidate geometric prediction residual information is determined based on the geometric distortion value and the first prediction residual code rate, and the geometric distortion value is used to characterize the geometry corresponding to the candidate geometric prediction residual information. Distortion, the first prediction residual code rate is used to represent the expected bit value for encoding the candidate geometric prediction residual information.
  • the at least two candidate geometric prediction residual information include candidate geometric prediction residual information related to the quantized geometric prediction residual information, and candidate geometric prediction residual information not related to the quantized geometric prediction residual information.
  • Prediction residual information Prediction residual information
  • Processor 610 is also used to:
  • target quantized geometric prediction residual information is candidate geometric prediction residual information related to the quantized geometric prediction residual information, based on the identifier corresponding to the target quantized geometric prediction residual information and the target quantization Entropy encoding of geometric prediction residual information;
  • target quantized geometric prediction residual information is candidate geometric prediction residual information unrelated to the quantized geometric prediction residual information
  • entropy encoding is performed based on an identifier corresponding to the target quantized geometric prediction residual information.
  • processor 610 is also used for:
  • Predictive encoding is performed on the geometric information of the point cloud to be encoded corresponding to the quantized point cloud obtained after deduplication processing.
  • processor 610 is also used for:
  • the geometric prediction residual information includes three-dimensional sub-geometric prediction residual information
  • the geometric quantization parameters include three sub-geometric quantization parameters respectively corresponding to the sub-geometric prediction residual information of the three dimensions.
  • processor 610 is also used for:
  • Predictive encoding is performed on the attribute information of the point cloud to be encoded to obtain attribute prediction residual information
  • Entropy coding is performed based on the quantized attribute prediction residual information to obtain an attribute code stream.
  • processor 610 is also used for:
  • Entropy coding is performed based on the target quantization attribute prediction residual information.
  • the target quantized attribute prediction residual information is the candidate attribute prediction residual information with the smallest rate-distortion cost among the at least two candidate attribute prediction residual information.
  • the rate-distortion cost corresponding to the candidate attribute prediction residual information is determined based on the attribute distortion value and the second prediction residual code rate, and the attribute distortion value is used to characterize the attribute corresponding to the candidate attribute prediction residual information Distortion, the second prediction residual code rate is used to represent the expected bit value for encoding the candidate attribute prediction residual information.
  • relevant candidate attribute prediction residual information, and candidate attribute prediction residual information not related to the quantized attribute prediction residual information are selected from the candidate attribute prediction residual information, and candidate attribute prediction residual information not related to the quantized attribute prediction residual information;
  • Processor 610 is also used to:
  • the target quantized attribute prediction residual information is candidate attribute prediction residual information related to the quantized attribute prediction residual information, based on the identifier corresponding to the target quantized geometric prediction residual information and the target quantization Attribute prediction residual information is entropy encoded;
  • target quantized attribute prediction residual information is candidate attribute prediction residual information unrelated to the quantized attribute prediction residual information
  • entropy coding is performed based on the identifier corresponding to the target quantized attribute prediction residual information.
  • the terminal in the embodiment of the present application can improve the rate control effect of the geometric code stream of the point cloud.
  • the terminal in the embodiment of the present application also includes: instructions or programs stored in the memory 609 and operable on the processor 610, and the processor 610 calls the instructions or programs in the memory 609 to execute the functions executed by the modules shown in FIG. method, and achieve the same technical effect, in order to avoid repetition, it is not repeated here.
  • the terminal is used to execute the point cloud decoding processing method:
  • the processor or the communication interface is configured to: perform entropy decoding on a geometric code stream to obtain quantized geometric prediction residual information; perform inverse quantization processing on the quantized geometric prediction residual information according to a geometric quantization parameter to obtain a geometric prediction residual difference information; performing predictive decoding based on the geometric prediction residual information to obtain geometric information of the point cloud to be decoded.
  • processor 610 is also used for:
  • the geometry quantization control parameter indicates that quantization processing is enabled, entropy decoding is performed on the geometry code stream to obtain quantized geometry prediction residual information.
  • processor 610 is also used for:
  • geometry quantization control parameter indicates that quantization processing is not enabled, entropy decoding is performed on the geometry code stream to obtain geometry prediction residual information.
  • processor 610 is also used for:
  • Inverse quantization processing is performed on the quantized geometric prediction residual information based on the second geometric quantization step size and a second preset geometric offset value.
  • the geometric prediction residual information includes three-dimensional sub-geometric prediction residual information
  • the geometric quantization parameters include three sub-geometric quantization parameters respectively corresponding to the sub-geometric prediction residual information of the three dimensions.
  • processor 610 is also used for:
  • Predictive decoding is performed based on the attribute prediction residual information to obtain attribute information of the point cloud to be decoded.
  • the terminal in the embodiment of the present application can improve the rate control effect of the geometric code stream of the point cloud.
  • the terminal in the embodiment of the present application also includes: instructions or programs stored in the memory 609 and operable on the processor 610, and the processor 610 calls the instructions or programs in the memory 609 to execute the functions executed by the modules shown in FIG. 11 method, and achieve the same technical effect, in order to avoid repetition, it is not repeated here.
  • the embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored, and when the program or instruction is executed by the processor, each process of the above-mentioned point cloud encoding processing method embodiment is realized, or, When the program or instruction is executed by the processor, each process of the above-mentioned point cloud decoding processing method embodiment can be achieved, and the same technical effect can be achieved. To avoid repetition, details are not repeated here.
  • the processor is the processor in the terminal described in the foregoing embodiments.
  • the readable storage medium includes computer readable storage medium, such as computer read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk, etc.
  • the embodiment of the present application further provides a chip, the chip includes a processor and a communication interface, the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the above point cloud encoding processing method
  • the chip includes a processor and a communication interface
  • the communication interface is coupled to the processor
  • the processor is used to run programs or instructions to implement the above point cloud encoding processing method
  • the chip mentioned in the embodiment of the present application may also be called a system-on-chip, a system-on-chip, a system-on-a-chip, or a system-on-a-chip.
  • the term “comprising”, “comprising” or any other variation thereof is intended to cover a non-exclusive inclusion such that a process, method, article or apparatus comprising a set of elements includes not only those elements, It also includes other elements not expressly listed, or elements inherent in the process, method, article, or device. Without further limitations, an element defined by the phrase “comprising a " does not preclude the presence of additional identical elements in the process, method, article, or apparatus comprising that element.
  • the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved. Functions are performed, for example, the described methods may be performed in an order different from that described, and various steps may also be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.
  • the methods of the above embodiments can be implemented by means of software plus a necessary general-purpose hardware platform, and of course also by hardware, but in many cases the former is better implementation.
  • the technical solution of the present application can be embodied in the form of a computer program product, which is stored in a storage medium (such as ROM/RAM, disk, etc.) , CD-ROM), including several instructions to make a terminal (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) execute the methods described in the various embodiments of the present application.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Compression Or Coding Systems Of Tv Signals (AREA)

Abstract

本申请公开了一种点云编码处理方法、点云解码处理方法及相关设备，属于点云处理技术领域，本申请实施例的点云编码处理方法，包括：基于待编码点云的几何信息进行预测编码，得到几何预测残差信息；依据几何量化参数对所述几何预测残差信息进行量化处理，得到量化几何预测残差信息；基于所述量化几何预测残差信息进行熵编码，得到几何码流。

Description

点云编码处理方法、点云解码处理方法及相关设备
相关申请的交叉引用
本申请主张在2021年6月11日在中国提交的中国专利申请No.202110656018.7的优先权，其全部内容通过引用包含于此。
技术领域
本申请属于点云处理技术领域，具体涉及一种点云编码处理方法、点云解码处理方法及相关设备。
背景技术
点云是三维物体或场景的一种表现形式，是由空间中一组无规则分布、表达三维物体或场景空间结构和表面属性的离散点集所构成。为了准确反映空间中的信息，所需离散点的数量相当大，而为了减少点云数据存储和传输时所占用的带宽，需要对点云数据进行编码压缩处理。点云数据通常由描述位置的几何信息如三维坐标(x，y，z)以及该位置的属性信息如颜色(R，G，B)或者反射率等构成。在点云编码压缩过程中对几何信息及属性信息的编码是分开进行的。
目前，在对点云的几何信息进行编码之前，会对几何信息进行取整和去重的预处理，而信源疏密程度对几何信息预处理的影响较大，例如，在信源分布较为稠密时，对几何信息进行预处理会急剧减少点云的数量，导致几何码流的速率受信源疏密程度的影响较大，从而导致点云的几何码流的速率控制效果较差。
发明内容
本申请实施例提供一种点云编码处理方法、点云解码处理方法及相关设备，能够解决几何码流的速率受信源疏密程度的影响较大，从而导致点云的几何码流的速率控制效果较差的问题。
第一方面，提供了一种点云编码处理方法，该方法包括：
基于待编码点云的几何信息进行预测编码，得到几何预测残差信息；
依据几何量化参数对所述几何预测残差信息进行量化处理，得到量化几何预测残差信息；
基于所述量化几何预测残差信息进行熵编码，得到几何码流。
第二方面，提供了一种点云解码处理方法，该方法包括：
对几何码流进行熵解码，得到量化几何预测残差信息；
依据几何量化参数对所述量化几何预测残差信息进行反量化处理，得到几何预测残差信息；
基于所述几何预测残差信息进行预测解码，得到待解码点云的几何信息。
第三方面，提供了一种点云编码处理装置，包括：
第一编码模块，用于基于待编码点云的几何信息进行预测编码，得到几何预测残差信息；
第一量化模块，用于依据几何量化参数对所述几何预测残差信息进行量化处理，得到量化几何预测残差信息；
第二编码模块，用于基于所述量化几何预测残差信息进行熵编码，得到几何码流。
第四方面，提供了一种点云解码处理装置，包括：
第一解码模块，用于对几何码流进行熵解码，得到量化几何预测残差信息；
第一反量化模块，用于依据几何量化参数对所述量化几何预测残差信息进行反量化处理，得到几何预测残差信息；
第二解码模块，用于基于所述几何预测残差信息进行预测解码，得到待解码点云的几何信息。
第五方面，提供了一种终端，该终端包括处理器、存储器及存储在所述存储器上并可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现如第一方面所述的方法的步骤；或者，所述程序或指令被所述处理器执行时实现如第二方面所述的方法的步骤。
第六方面，提供了一种终端，包括处理器及通信接口，其中，所述处理器或者所述通信接口用于：
基于待编码点云的几何信息进行预测编码，得到几何预测残差信息；
依据几何量化参数对所述几何预测残差信息进行量化处理，得到量化几何预测残差信息；
基于所述量化几何预测残差信息进行熵编码，得到几何码流。
第七方面，提供了一种终端，包括处理器及通信接口，其中，所述处理器或者所述通信接口用于：
对几何码流进行熵解码，得到量化几何预测残差信息；
依据几何量化参数对所述量化几何预测残差信息进行反量化处理，得到几何预测残差信息；
基于所述几何预测残差信息进行预测解码，得到待解码点云的几何信息。
第八方面，提供了一种可读存储介质，所述可读存储介质上存储程序或指令，所述程序或指令被处理器执行时实现如第一方面所述的点云编码处理方法的步骤，或者，所述程序或指令被处理器执行时实现如第二方面所述的点云解码处理方法的步骤。
第九方面，提供了一种芯片，所述芯片包括处理器和通信接口，所述通信接口和所述处理器耦合，所述处理器用于运行程序或指令，实现如第一方面所述的点云编码处理方法的步骤，或者，实现如第二方面所述的点云解码处理方法的步骤。
第十方面，提供了一种计算机程序/程序产品，所述计算机程序/程序产品被存储在非易失的存储介质中，所述程序/程序产品被至少一个处理器执行以实现如第一方面所述的点云编码处理方法的步骤，或者，实现如第二方面所述的点云解码处理方法的步骤。。
第十一方面，提供了一种通信设备，被配置为执行如第一方面所述的点云编码处理方法的步骤，或，执行如第二方面任一项所述的点云解码处理方法的步骤。
本申请实施例中，基于待编码点云的几何信息进行预测编码，得到几何预测残差信息；依据几何量化参数对所述几何预测残差信息进行量化处理，得到量化几何预测残差信息；基于所述量化几何预测残差信息进行熵编码，得到几何码流。这样，通过几何量化参数对所述几何预测残差信息进行量化 处理，降低了信源疏密程度对几何码流速率的影响，能够提高点云的几何码流的速率控制效果。
附图说明
图1是一种点云AVS编码器框架示意图之一；
图2是一种点云AVS解码器框架示意图之一；
图3是本申请实施例提供的一种点云编码处理方法的流程图；
图4是一种点云AVS编码器框架示意图之二；
图5是一种点云AVS解码器框架示意图之二；
图6是本申请实施例提供的一种点云解码处理方法的流程图；
图7是本申请实施例提供的一种点云编码处理装置的结构图之一；
图8是本申请实施例提供的一种点云编码处理装置的结构图之二；
图9是本申请实施例提供的一种点云编码处理装置的结构图之三；
图10是本申请实施例提供的一种点云编码处理装置的结构图之四；
图11是本申请实施例提供的一种点云解码处理装置的结构图之一；
图12是本申请实施例提供的一种点云解码处理装置的结构图之二；
图13是本申请实施例提供的一种通信设备的结构图；
图14是本申请实施例提供的一种终端的结构图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚描述，显然，所描述的实施例是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员所获得的所有其他实施例，都属于本申请保护的范围。
本申请的说明书和权利要求书中的术语“第一”、“第二”等是用于区别类似的对象，而不用于描述特定的顺序或先后次序。应该理解这样使用的术语在适当情况下可以互换，以便本申请的实施例能够以除了在这里图示或描述的那些以外的顺序实施，且“第一”、“第二”所区别的对象通常为一类，并不限定对象的个数，例如第一对象可以是一个，也可以是多个。此外，说 明书以及权利要求中“和/或”表示所连接对象的至少其中之一，字符“/”一般表示前后关联对象是一种“或”的关系。
本申请实施例中的编解码方法对应的编解码端可以为终端，该终端也可以称作终端设备或者用户终端(User Equipment，UE)，终端可以是手机、平板电脑(Tablet Personal Computer)、膝上型电脑(Laptop Computer)或称为笔记本电脑、个人数字助理(Personal Digital Assistant，PDA)、掌上电脑、上网本、超级移动个人计算机(ultra-mobile personal computer，UMPC)、移动上网装置(Mobile Internet Device，MID)、增强现实(augmented reality，AR)/虚拟现实(virtual reality，VR)设备、机器人、可穿戴式设备(Wearable Device)或车载设备(VUE)、行人终端(PUE)等终端侧设备，可穿戴式设备包括：智能手表、手环、耳机、眼镜等。需要说明的是，在本申请实施例并不限定终端的具体类型。
为了方便理解，以下对本申请实施例涉及的一些内容进行说明：
如图1所示，在点云数字音视频编解码技术标准(Audio Video coding Standard，AVS)编码器框架中，点云的几何信息和属性信息是分开编码的。首先对几何信息进行坐标转换，使点云全部包含在一个包围盒(bounding box)中，然后再进行坐标量化。量化主要起到缩放的作用，由于量化会对几何坐标取整，使得一部分点的几何信息相同，称为重复点，根据参数来决定是否移除重复点，量化和移除重复点这两个步骤又被称为体素化过程。接下来，对包围盒进行多叉树划分，例如八叉树、四叉树或二叉树划分。在基于多叉树的几何信息编码框架中，将包围盒八等分为8个子立方体，对非空的的子立方体继续进行划分，直到划分得到叶子节点为1x1x1的单位立方体时停止划分，对叶子结点中的点数进行编码，生成二进制码流。目前AVS几何划分顺序包括两种：
1、广度优先遍历顺序：对几何进行八叉树划分时，首先对当前同一层的节点进行划分，直至划分完当前层上的所有节点，才会继续划分下一层的节点，最终当划分得到的叶子结点为1x1x1的单位立方体时停止划分。
2、深度优先遍历顺序：对几何进行八叉树划分时，首先会对当前层的第一个节点进行不断地划分，直到划分得到的叶子结点为1x1x1的单位立方体 时停止划分当前节点。按照该顺序，对当前层后续的节点进行划分，直至当前层上的节点划分完成停止。
几何编码完成后，对几何信息进行重建，用于后面的重着色。属性编码主要针对的是颜色和反射率信息。首先根据参数判断是否进行颜色空间转换，若进行颜色空间转换，则将颜色信息从红绿蓝(Red Green Blue，RGB)颜色空间转换到亮度色彩(YUV)颜色空间。然后，利用原始点云对几何重建点云进行重着色，使得未编码的属性信息与重建的几何信息对应起来。在颜色信息编码中，通过莫顿码对点云进行排序后，利用几何空间关系搜索待预测点的最近邻，并利用所找到邻居的重建属性值对待预测点进行预测得到预测属性值，然后将真实属性值和预测属性值进行差分得到预测残差，最后对预测残差进行量化并编码，生成二进制码流。
可选地，AVS解码流程与编码流程对应，具体的，AVS解码器框架如图2所示。
需要说明的是，AVS编码框架可以包括预处理及编码两个阶段，在预处理阶段对点云的处理可以称为环外处理，而在预处理完成后，即编码阶段对点云的处理可以称为环内处理。
下面结合附图，通过一些实施例及其应用场景对本申请实施例提供的点云编码处理方法进行详细地说明。
参见图3，图3是本申请实施例提供的一种点云编码处理方法的流程图，如图3所示，点云编码处理方法包括以下步骤：
步骤101、基于待编码点云的几何信息进行预测编码，得到几何预测残差信息。
其中，几何信息可以包括几何位置。在进行预测编码时，可以对待编码点云的几何信息建立预测候选列表，从预测候选列表中选取最佳的几何预测值，将最佳的几何预测值与几何信息做差得到几何预测残差信息。预测候选列表中的每个几何预测值均可以与一个几何预测模式对应。示例地，可以预先建立预测候选列表，该预测候选列表可以包括N个几何预测值，其中，N个几何预测值与N个几何预测模式一一对应，N为大于1的正整数。示例性的，若N的数量为4，即预测候选列表包括4个几何预测值，待编码点云为 所有点云中的第5个待编码的点云，则可以利用位于待编码点云之前的编码顺序为1至4的4个待编码点云的几何信息，确定几何预测值。例如，几何预测值的确定规则可以是，第一个几何预测值为4个待编码点云的几何信息的和；第二个几何预测值为4个待编码点云的最小几何信息；第三个几何预测值为4个待编码点云的几何信息的平均值；第四个几何预测值为第4个待编码点云的几何信息与第3个待编码点云的几何信息的差值。其中，待编码点云的几何信息可以表征为待编码点云的三维坐标(x，y，z)。
应理解，关于几何预测值具体的确定规则可以灵活设定，本实施例在此不做具体限定。
步骤102、依据几何量化参数对所述几何预测残差信息进行量化处理，得到量化几何预测残差信息。
其中，可以确定几何量化控制参数是否指示启用量化处理；在所述几何量化控制参数指示启用量化处理的情况下，依据几何量化参数对所述几何预测残差信息进行量化处理，得到量化几何预测残差信息，基于所述量化几何预测残差信息进行熵编码，得到几何码流；在所述几何量化控制参数指示不启用量化处理的情况下，依据所述几何预测残差信息进行熵编码，得到几何码流。
另外，几何量化参数可以从配置文件中读取，示例地，可以读取配置文件中参数GeomQP的值作为几何量化参数。
步骤103、基于所述量化几何预测残差信息进行熵编码，得到几何码流。
其中，可以对量化几何预测残差信息和几何预测模式进行熵编码，得到几何码流。
需要说明的是，在基于八叉树的几何编码方案中，会在预处理中进行对几何信息有损量化的过程，在预处理中通过量化步长对原始点云数据进行去重和取整的有损量化，不仅对原始点云数据进行取整，且在取整后去除重复点，实现有损量化。该预处理中的有损量化会存在以下问题：当信源分布较为稀疏、均匀时，同过设置不同的量化步长可以均匀地减少量化后的点数，但当信源分布较为稠密、集中时，较小的量化步长就会导致点数的急剧减少；有损量化在环外完成，通过量化步长实现，可以认为是对原始点云数据进行 了一次下采样操作，而下采样的点数受到信源的影响，因此量化后的几何码流也会受到信源的影响，同时在点数固定的情况下无法对点的坐标进行进一步的有损量化，从而无法做到几何码流的速率控制；预处理后的点云受到信源的稠密程度的影响较大，在信源未知的情况下，无法通过调整环外的量化步长来准确控制预处理后的点云的几何信息的质量。
本申请实施例中，通过几何量化参数对所述几何预测残差信息进行量化处理，在点云编码中引入环内有损量化，降低了信源疏密程度对几何码流速率的影响，能够提高点云的几何码流的速率控制效果。
本申请实施例中，基于待编码点云的几何信息进行预测编码，得到几何预测残差信息；依据几何量化参数对所述几何预测残差信息进行量化处理，得到量化几何预测残差信息；基于所述量化几何预测残差信息进行熵编码，得到几何码流。这样，通过几何量化参数对所述几何预测残差信息进行量化处理，降低了信源疏密程度对几何码流速率的影响，能够提高点云的几何码流的速率控制效果。
可选的，所述依据几何量化参数对所述几何预测残差信息进行量化处理，包括：
确定几何量化控制参数是否指示启用量化处理；
在所述几何量化控制参数指示启用量化处理的情况下，依据几何量化参数对所述几何预测残差信息进行量化处理。
其中，几何量化控制参数指示启用量化处理，可以认为是指示对几何预测残差信息启用环内有损量化。几何量化控制参数可以从配置文件中读取，示例地，可以读取配置文件中参数geometry_enable_quantizated_flag的值作为几何量化控制参数。参数geometry_enable_quantizated_flag可以为gps(geometry parameters set)高层语法元素中新引入的参数。在几何量化控制参数被配置为1时，可以表示启用量化处理；在几何量化控制参数被配置为0时，可以表示不启用量化处理。
该实施方式中，通过几何量化控制参数确定是否依据几何量化参数对所述几何预测残差信息进行量化处理，能够提高点云的几何信息编码的灵活性。
可选的，所述确定几何量化控制参数是否指示启用量化处理之后，所述 方法还包括：
在所述几何量化控制参数指示不启用量化处理的情况下，依据所述几何预测残差信息进行熵编码，得到几何码流。
其中，几何量化控制参数指示不启用量化处理，可以认为是指示对几何预测残差信息不启用环内有损量化。
该实施方式中，在所述几何量化控制参数指示不启用量化处理的情况下，依据所述几何预测残差信息进行熵编码，从而可以依据几何量化控制参数确定在对几何信息的预测编码过程中是否对几何预测残差信息进行量化处理，能够提高点云的几何信息编码的灵活性。
可选的，所述对待编码点云的几何信息进行预测编码，包括：
基于待编码点云对应的节点标识将所述待编码点云划分为第一子待编码点云和第二子待编码点云；
在几何编码控制参数指示第一编码模式的情况下，对所述第一子待编码点云的几何信息进行预测编码；
在所述几何编码控制参数指示第二编码模式的情况下，对所述第二子待编码点云的几何信息进行预测编码。
其中，可以依据待编码点云对应的节点标识与预设阈值的关系将待编码点云划分为第一子待编码点云和第二子待编码点云。第一子待编码点云可以为低比特待编码点云，第二子待编码点云可以为高比特待编码点云。高比特待编码点云的几何信息可以包括八叉树高比特坐标，低比特待编码点云的几何信息可以包括八叉树低比特坐标。
示例地，如图4所示，对于高比特待编码点云，进行八叉树构建实现八叉树编码；对于低比特待编码点云，进行几何预测和残差量化，实现预测编码。而在解码端，如图5所示，通过八叉树重建，得到高比特待编码点云；通过逆量化和几何重建，得到低比特待编码点云。
以待编码点云的几何信息用莫顿码表示，且通过莫顿码对几何信息构建几何八叉树为例，待编码点云对应的节点标识可以为八叉树编码过程中的编码层的层数。示例地，预设阈值可以为5，全部待编码点云可以包括10个编码层，可以将第1个编码层至第4个编码层对应的待编码点云作为高比特待 编码点云，将第5个编码层至第10个编码层对应的待编码点云作为低比特待编码点云。
另外，预设阈值可以为参数octree_division_end_nodeSizeLog2[3]的值。可以从配置文件读取参数octree_division_end_nodeSizeLog2[3]的值作为预设阈值。在几何量化参数大于或等于预设阈值时，几何预测残差信息均量化为0，可以不需要对量化几何预测残差信息进行熵编码。在第一编码模式且预设阈值与预处理中环外量化的第一量化参数相匹配时，点云的有损量化与现有的量化保持一致。在几何量化参数小于预设阈值时，几何预测残差信息不会量化为0，可以基于所述量化几何预测残差信息进行熵编码。
该实施方式中，在几何编码控制参数指示不同的编码模式的情况下，对几何信息进行预测编码的待编码点云不同，用户可以通过设置几何编码控制参数修改编码模式，从而修改待编码点云的编码方式，从而能够提高点云编码的灵活性。
可选的，所述基于所述量化几何预测残差信息进行熵编码，包括：
基于所述量化几何预测残差信息确定至少两个候选几何预测残差信息；
获取所述至少两个候选几何预测残差信息对应的率失真代价；
依据所述至少两个候选几何预测残差信息对应的率失真代价确定目标量化几何预测残差信息；
基于所述目标量化几何预测残差信息进行熵编码。
其中，所述至少两个候选几何预测残差信息中可以包括与所述量化几何预测残差信息相关的候选几何预测残差信息，及与所述量化几何预测残差信息不相关的候选几何预测残差信息。示例地，所述至少两个候选几何预测残差信息可以包括量化几何预测残差信息及固定值{0，0，0}。
该实施方式中，针对量化几何预测残差信息，引入率失真优化算法处理得到目标量化几何预测残差信息，基于所述目标量化几何预测残差信息进行熵编码，能够提升对几何信息有损编码的效率。
可选的，所述目标量化几何预测残差信息为所述至少两个候选几何预测残差信息中率失真代价最小的候选几何预测残差信息。
其中，可以以候选列表的形式存储所述至少两个候选几何预测残差信息， 将候选列表中的第一个候选几何预测残差信息作为最佳候选几何预测残差信息；遍历候选列表中的候选几何预测残差信息；若当前候选几何预测残差信息对应的率失真代价小于最佳候选几何预测残差信息对应的率失真代价，则将当前候选几何预测残差信息更新为最佳候选几何预测残差信息，否则，不对最佳候选几何预测残差信息进行更新；在遍历完候选列表后，将最佳候选几何预测残差信息确定为目标量化几何预测残差信息。在确定目标量化几何预测残差信息之后，可以将目标量化几何预测残差信息输入编码器进行熵编码。
该实施方式中，将所述至少两个候选几何预测残差信息中率失真代价最小的候选几何预测残差信息确定为目标量化几何预测残差信息，从而能够优化几何信息的有损编码过程，提高点云编码效率。
可选的，所述候选几何预测残差信息对应的率失真代价基于几何失真值及第一预测残差码率确定，所述几何失真值用于表征所述候选几何预测残差信息对应的几何失真，所述第一预测残差码率用于表征编码所述候选几何预测残差信息预计的比特值。
其中，所述候选几何预测残差信息对应的率失真代价可以与几何失真值及第一预测残差码率均正相关。示例地，所述候选几何预测残差信息对应的率失真代价cost1可以为：
cost 1＝dist1+λ1*rate1
其中，λ1可以表示码率与失真在率失真代价中的权重参数，示例地，λ1可以设置为0.4，0.5或者0.6等等；rate1可以表示第一预测残差码率；dist可以表示几何失真值。几何失真值dist1的计算公式可以如下所示：
dist1＝normal1(recPos-oriPos)
其中，函数normal1表示求取表达式的一范数，recPos表示利用候选几何预测残差信息与几何预测值重建的几何坐标，oriPos表示原始几何坐标。
该实施方式中，所述候选几何预测残差信息对应的率失真代价基于几何失真值及第一预测残差码率确定，能够较为准确地确定候选几何预测残差信息对应的率失真代价。
可选的，所述至少两个候选几何预测残差信息中包括与所述量化几何预 测残差信息相关的候选几何预测残差信息，及与所述量化几何预测残差信息不相关的候选几何预测残差信息；
所述基于所述目标量化几何预测残差信息进行熵编码，包括：
在所述目标量化几何预测残差信息为与所述量化几何预测残差信息相关的候选几何预测残差信息的情况下，基于所述目标量化几何预测残差信息对应的标识和所述目标量化几何预测残差信息进行熵编码；
在所述目标量化几何预测残差信息为与所述量化几何预测残差信息不相关的候选几何预测残差信息的情况下，基于所述目标量化几何预测残差信息对应的标识进行熵编码。
其中，每个候选几何预测残差信息均可以对应设置有一个标识。候选几何预测残差信息与所述量化几何预测残差信息相关，可以是，候选几何预测残差信息可以基于所述量化几何预测残差信息获得，示例地，候选几何预测残差信息等于量化几何预测残差信息，该候选几何预测残差信息对应的标识可以为1；或者，候选几何预测残差信息等于量化几何预测残差信息的整数倍，该候选几何预测残差信息对应的标识可以为2，等等；候选几何预测残差信息与所述量化几何预测残差信息不相关，可以是，候选几何预测残差信息为预设几何预测残差信息，示例地，可以为(0，0，0)，该候选几何预测残差信息对应的标识可以为0。
另外，可以设置有几何率失真优化控制参数，若几何率失真优化控制参数为第一预设值，则所述基于所述目标量化几何预测残差信息进行熵编码，包括：在所述目标量化几何预测残差信息为与所述量化几何预测残差信息相关的候选几何预测残差信息的情况下，基于所述目标量化几何预测残差信息对应的标识和所述目标量化几何预测残差信息进行熵编码；在所述目标量化几何预测残差信息为与所述量化几何预测残差信息不相关的候选几何预测残差信息的情况下，基于所述目标量化几何预测残差信息对应的标识进行熵编码。
若几何率失真优化控制参数为第二预设值，则所述基于所述目标量化几何预测残差信息进行熵编码，包括：在所述目标量化几何预测残差信息为与所述量化几何预测残差信息相关的候选几何预测残差信息的情况下，基于所 述目标量化几何预测残差信息进行熵编码；在所述目标量化几何预测残差信息为与所述量化几何预测残差信息不相关的候选几何预测残差信息的情况下，基于所述目标量化几何预测残差信息进行熵编码。
本实施例对第一预设值和第二预设值不进行限定。示例地，第一预设值可以为1，第二预设值可以为0。
进一步的，可以通过目标量化几何预测残差信息对应的标识确定所述目标量化几何预测残差信息与所述量化几何预测残差信息相关或者不相关。在解码端进行解码时，可以先解析目标量化几何预测残差信息对应的标识，若根据目标量化几何预测残差信息对应的标识确定所述目标量化几何预测残差信息与所述量化几何预测残差信息不相关，则可以根据目标量化几何预测残差信息对应的标识查找到目标量化几何预测残差信息；若根据目标量化几何预测残差信息对应的标识确定所述目标量化几何预测残差信息与所述量化几何预测残差信息相关，则可以从几何码流中解码得到目标量化几何预测残差信息。
该实施方式中，在所述目标量化几何预测残差信息为与所述量化几何预测残差信息相关的候选几何预测残差信息的情况下，基于所述目标量化几何预测残差信息对应的标识和所述目标量化几何预测残差信息进行熵编码；在所述目标量化几何预测残差信息为与所述量化几何预测残差信息不相关的候选几何预测残差信息的情况下，基于所述目标量化几何预测残差信息对应的标识进行熵编码。这样，对于部分目标量化几何预测残差信息，可以不对其进行编码，而仅编码目标量化几何预测残差信息对应的标识，能够进一步提高编码效率。
可选的，所述对待编码点云的几何信息进行预测编码，包括：
依据预先设置的第一量化步长获取待编码点云对应的量化点云；
对所述量化点云进行去重处理；
将去重处理后得到的量化点云对应的待编码点云的几何信息进行预测编码。
其中，所述依据预先设置的第一量化步长获取待编码点云对应的量化点云之前，如图4所示，可以对待编码点云进行坐标平移处理，坐标平移处理 可以将包围盒移动至坐标原点(0，0，0)，其中包围盒表示包含输入点云中所有点的最小长方体。第一量化步长可以由用户预先设定，示例地，第一量化步长QS可以为：
QS＝2 i
其中，在无损情况下，i＝0，在有损情况下，根据不同的量化等级(r01，…，r06)，i＝9、8、6、5、3、2，即QS的值可以为512，256，64，32，8，4，1。
待编码点云对应的量化点云的获取方式可以如下：
X＝round(x/QS)
Y＝round(y/QS)
Z＝round(z/QS)
其中，(X，Y，Z)表示量化点云的量化坐标，(x，y，z)表示原始点云的坐标，QS表示第一量化步长，round(s)函数表示返回距s最近的整数，该函数的具体定义可以如下所示：
Figure PCTCN2022097635-appb-000001
在计算得到量化点云的量化坐标后，会存在多个原始点云的量化坐标相同的情况，具有相同量化坐标的原始点云为重复点。可以在去重参数geom_remove_dup_flag指示去除重复点的情况下，对所述量化点云进行去重处理。去重处理后得到的量化点云对应的待编码点云坐标为原始点云坐标，并未对点云坐标进行量化操作。
该实施方式中，将去重处理后得到的量化点云对应的待编码点云的几何信息进行预测编码，从而在环外量化时仅对点云数量进行下采样，而对点云坐标不进行量化。
可选的，所述依据几何量化参数对所述几何预测残差信息进行量化处理，包括：
依据几何量化参数确定第一几何量化步长；
基于所述第一几何量化步长及第一预设几何偏移值对所述几何预测残差信息进行量化处理。
其中，第一几何量化步长QS1可以为：
Figure PCTCN2022097635-appb-000002
其中，2 shift1可以表示第一预设几何偏移值，shift1可以表示量化处理过程中偏移的比特位数，shift1越大则表征量化结果越精确，QP1可以表示几何量化参数。
示例地，shift1可以配置为14。
进行量化处理得到的量化几何预测残差信息QtRes1可以为：
Figure PCTCN2022097635-appb-000003
其中，Res1可以表示几何预测残差信息，offset1可以表示第一预设几何偏移值的一半，即offset1为2 shift1-1，通过offset1可以实现四舍五入的操作。
该实施方式中，依据几何量化参数确定第一几何量化步长，基于所述第一几何量化步长及第一预设几何偏移值对所述几何预测残差信息进行量化处理，能够获得较好的量化效果。
可选的，所述几何预测残差信息包括三个维度的子几何预测残差信息；
所述几何量化参数包括与所述三个维度的子几何预测残差信息分别对应的三个子几何量化参数。
其中，三个维度可以分别为三维坐标系中的X、Y、Z维度。可以通过配置文件(cfg)中的参数GeomQP[3]来配置三个子几何量化参数，该三个子几何量化参数可以分别对三个维度的子几何预测残差信息进行对应的量化。
该实施方式中，所述几何量化参数包括与所述三个维度的子几何预测残差信息分别对应的三个子几何量化参数，能够分别对三个维度的子几何预测残差信息进行量化，从而能够提高几何信息环内有损量化的鲁棒性和适应性。
可选的，所述方法还包括：
对所述待编码点云的属性信息进行预测编码，得到属性预测残差信息；
依据属性量化参数对所述属性预测残差信息进行量化处理，得到量化属性预测残差信息；
基于所述量化属性预测残差信息进行熵编码，得到属性码流。
其中，可以依据属性量化参数确定第一属性量化步长；基于所述第一属性量化步长及预设属性偏移值对所述属性预测残差信息进行量化处理。
其中，第一属性量化步长QS2可以为：
Figure PCTCN2022097635-appb-000004
其中，QP2可以表示属性量化参数。
进行量化处理得到的量化属性预测残差信息QtRes2可以为：
Figure PCTCN2022097635-appb-000005
其中，Res2可以表示属性预测残差信息，offset2可以表示预设属性偏移值，示例地，offset2可以设置为0.5。
另外，在进行预测编码时，可以对待编码点云的属性信息建立预测候选列表，从预测候选列表中选取最佳的属性预测值，将最佳的属性预测值与属性信息做差得到属性预测残差信息。预测候选列表中的每个属性预测值均可以与一个属性预测模式对应。示例地，可以预先建立预测候选列表，该预测候选列表可以包括N个属性预测值，其中，N个属性预测值与N个属性预测模式一一对应，N为大于1的正整数。示例性的，若N的数量为4，即预测候选列表包括4个属性预测值，待编码点云为所有点云中的第5个待编码的点云，则可以利用位于待编码点云之前的编码顺序为1至4的4个待编码点云的属性信息，确定属性预测值。例如，属性预测值的确定规则可以是，第一个属性预测值为4个待编码点云的属性信息的和；第二个属性预测值为4个待编码点云的最小属性信息；第三个属性预测值为4个待编码点云的属性信息的平均值；第四个属性预测值为第4个待编码点云的属性信息与第3个待编码点云的属性信息的差值。其中，待编码点云的属性信息可以表征为待编码点云的三维坐标(x，y，z)。
应理解，关于属性预测值具体的确定规则可以灵活设定，本实施例在此不做具体限定。
该实施方式中，对所述待编码点云的属性信息进行预测编码，得到属性预测残差信息；依据属性量化参数对所述属性预测残差信息进行量化处理，得到量化属性预测残差信息；基于所述量化属性预测残差信息进行熵编码， 得到属性码流。这样，在属性信息编码的过程中引入量化处理，能够提高属性信息编码效率。
可选的，所述基于所述量化属性预测残差信息进行熵编码，包括：
基于所述量化属性预测残差信息确定至少两个候选属性预测残差信息；
获取所述至少两个候选属性预测残差信息对应的率失真代价；
依据所述至少两个候选属性预测残差信息对应的率失真代价确定目标量化属性预测残差信息；
基于所述目标量化属性预测残差信息进行熵编码。
其中，所述至少两个候选属性预测残差信息中可以包括与所述量化属性预测残差信息相关的候选属性预测残差信息，及与所述量化属性预测残差信息不相关的候选属性预测残差信息。示例地，在编码颜色时，所述至少两个候选属性预测残差信息可以包括量化属性预测残差信息及固定值{0，0，0}。
该实施方式中，针对量化属性预测残差信息，引入率失真优化算法处理得到目标量化属性预测残差信息，基于所述目标量化属性预测残差信息进行熵编码，能够提升对属性信息有损编码的效率。
可选的，所述目标量化属性预测残差信息为所述至少两个候选属性预测残差信息中率失真代价最小的候选属性预测残差信息。
其中，可以以候选列表的形式存储所述至少两个候选属性预测残差信息，将候选列表中的第一个候选属性预测残差信息作为最佳候选属性预测残差信息；遍历候选列表中的候选属性预测残差信息；若当前候选属性预测残差信息对应的率失真代价小于最佳候选属性预测残差信息对应的率失真代价，则将当前候选属性预测残差信息更新为最佳候选属性预测残差信息，否则，不对最佳候选属性预测残差信息进行更新；在遍历完候选列表后，将最佳候选属性预测残差信息确定为目标量化属性预测残差信息。在确定目标量化属性预测残差信息之后，可以将目标量化属性预测残差信息输入编码器进行熵编码。
该实施方式中，将所述至少两个候选属性预测残差信息中率失真代价最小的候选属性预测残差信息确定为目标量化属性预测残差信息，从而能够优化属性信息的有损编码过程，提高点云编码效率。
可选的，所述候选属性预测残差信息对应的率失真代价基于属性失真值及第二预测残差码率确定，所述属性失真值用于表征所述候选属性预测残差信息对应的属性失真，所述第二预测残差码率用于表征编码所述候选属性预测残差信息预计的比特值。
其中，所述候选属性预测残差信息对应的率失真代价可以与属性失真值及第二预测残差码率均正相关。示例地，所述候选属性预测残差信息对应的率失真代价cost2可以为：
cost2＝dist2+λ2*rate2
其中，λ2可以表示码率与失真在率失真代价中的权重参数，示例地，λ2可以设置为0.4，0.5或者0.6等等；rate2可以表示第二预测残差码率；dist2可以表示属性失真值。属性失真值dist2的计算公式可以如下所示：
dist2＝normal1(recAttri-oriAttri)
其中，函数normal1表示求取表达式的一范数，recAttri表示利用候选属性预测残差信息与属性预测值得到的重建属性值，oriAttri表示原始属性值。
该实施方式中，所述候选属性预测残差信息对应的率失真代价基于属性失真值及第二预测残差码率确定，能够较为准确地确定候选属性预测残差信息对应的率失真代价。
可选的，所述至少两个候选属性预测残差信息中包括与所述量化属性预测残差信息相关的候选属性预测残差信息，及与所述量化属性预测残差信息不相关的候选属性预测残差信息；
所述基于所述目标量化属性预测残差信息进行熵编码，包括：
在所述目标量化属性预测残差信息为与所述量化属性预测残差信息相关的候选属性预测残差信息的情况下，基于所述目标量化几何预测残差信息对应的标识和所述目标量化属性预测残差信息进行熵编码；
在所述目标量化属性预测残差信息为与所述量化属性预测残差信息不相关的候选属性预测残差信息的情况下，基于所述目标量化属性预测残差信息对应的标识进行熵编码。
其中，候选属性预测残差信息与所述量化属性预测残差信息相关，可以是，候选属性预测残差信息可以基于所述量化属性预测残差信息获得，示例 地，候选属性预测残差信息等于量化属性预测残差信息，或者，候选属性预测残差信息等于量化属性预测残差信息的整数倍等等；候选属性预测残差信息与所述量化属性预测残差信息不相关，可以是，候选属性预测残差信息为预设属性预测残差信息，示例地，可以为(0，0，0)。
另外，可以设置有属性率失真优化控制参数，若属性率失真优化控制参数为第三预设值，则所述基于所述目标量化属性预测残差信息进行熵编码，包括：在所述目标量化属性预测残差信息为与所述量化属性预测残差信息相关的候选属性预测残差信息的情况下，基于所述目标量化属性预测残差信息对应的标识和所述目标量化属性预测残差信息进行熵编码；在所述目标量化属性预测残差信息为与所述量化属性预测残差信息不相关的候选属性预测残差信息的情况下，基于所述目标量化属性预测残差信息对应的标识进行熵编码。
若属性率失真优化控制参数为第四预设值，则所述基于所述目标量化属性预测残差信息进行熵编码，包括：在所述目标量化属性预测残差信息为与所述量化属性预测残差信息相关的候选属性预测残差信息的情况下，基于所述目标量化属性预测残差信息进行熵编码；在所述目标量化属性预测残差信息为与所述量化属性预测残差信息不相关的候选属性预测残差信息的情况下，基于所述目标量化属性预测残差信息进行熵编码。
本实施例对第三预设值和第四预设值不进行限定。示例地，第三预设值可以为1，第四预设值可以为0。
进一步的，可以通过目标量化属性预测残差信息对应的标识确定所述目标量化属性预测残差信息与所述量化属性预测残差信息相关或者不相关。在解码端进行解码时，可以先解析目标量化属性预测残差信息对应的标识，若根据目标量化属性预测残差信息对应的标识确定所述目标量化属性预测残差信息与所述量化属性预测残差信息不相关，则可以根据目标量化属性预测残差信息对应的标识查找到目标量化属性预测残差信息；若根据目标量化属性预测残差信息对应的标识确定所述目标量化属性预测残差信息与所述量化属性预测残差信息相关，则可以从属性码流中解码得到目标量化属性预测残差信息。
该实施方式中，在所述目标量化属性预测残差信息为与所述量化属性预测残差信息相关的候选属性预测残差信息的情况下，基于所述目标量化属性预测残差信息对应的标识和所述目标量化属性预测残差信息进行熵编码；在所述目标量化属性预测残差信息为与所述量化属性预测残差信息不相关的候选属性预测残差信息的情况下，基于所述目标量化属性预测残差信息对应的标识进行熵编码。这样，对于部分目标量化属性预测残差信息，可以不对其进行编码，而仅编码目标量化属性预测残差信息对应的标识，能够进一步提高编码效率。
参见图，图6是本申请实施例提供的一种点云解码处理方法的流程图，如图6所示，点云解码处理方法包括以下步骤：
步骤201、对几何码流进行熵解码，得到量化几何预测残差信息；
步骤202、依据几何量化参数对所述量化几何预测残差信息进行反量化处理，得到几何预测残差信息；
步骤203、基于所述几何预测残差信息进行预测解码，得到待解码点云的几何信息。
其中，可以对几何码流进行熵解码，得到量化几何预测残差信息和几何预测模式。可以基于所述几何预测残差信息和几何预测模式进行预测解码，得到待解码点云的几何信息。示例地，可以对几何预测模式进行解析，根据几何预测模式选取对应的几何预测值；将几何预测值与几何预测残差信息相加，得到待解码点云的几何信息。几何信息可以包括几何坐标。
可选的，所述对所述几何码流进行熵解码，得到量化几何预测残差信息，包括：
确定几何量化控制参数是否指示启用量化处理；
在所述几何量化控制参数指示启用量化处理的情况下，对所述几何码流进行熵解码，得到量化几何预测残差信息。
可选的，所述确定几何量化控制参数是否指示启用量化处理之后，所述方法还包括：
在所述几何量化控制参数指示不启用量化处理的情况下，对所述几何码流进行熵解码，得到几何预测残差信息。
可选的，所述依据几何量化参数对所述量化几何预测残差信息进行反量化处理，包括：
依据几何量化参数确定第二几何量化步长；
基于所述第二几何量化步长及第二预设几何偏移值对所述量化几何预测残差信息进行反量化处理。
其中，第二几何量化步长QS3可以为：
Figure PCTCN2022097635-appb-000006
其中，2 shift3可以表示第二预设几何偏移值，shift3可以表示量化处理过程中偏移的比特位数，shift3越大则表征量化结果越精确，QP1可以表示几何量化参数。
示例地，shift3可以配置为6。
进行反量化处理得到的几何预测残差信息RQtRes1可以为：
Figure PCTCN2022097635-appb-000007
其中，QtRes1可以表示量化几何预测残差信息，offset3可以表示第二预设几何偏移值的一半，即offset3为2 shift3-1。
可选的，所述几何预测残差信息包括三个维度的子几何预测残差信息；
所述几何量化参数包括与所述三个维度的子几何预测残差信息分别对应的三个子几何量化参数。
可选的，所述方法还包括：
对属性码流进行熵解码，得到量化属性预测残差信息；
依据属性量化参数对所述量化属性预测残差信息进行反量化处理，得到属性预测残差信息；
基于所述属性预测残差信息进行预测解码，得到所述待解码点云的属性信息。
其中，所述对属性码流进行熵解码，得到量化属性预测残差信息，可以包括：确定属性量化控制参数是否指示启用量化处理；在所述属性量化控制参数指示启用量化处理的情况下，对所述属性码流进行熵解码，得到量化属性预测残差信息；在所述属性量化控制参数指示不启用量化处理的情况下， 对所述属性码流进行熵解码，得到属性预测残差信息。
其中，可以对属性码流进行熵解码，得到量化属性预测残差信息和属性预测模式。可以基于所述属性预测残差信息和属性预测模式进行预测解码，得到待解码点云的属性信息。示例地，可以对属性预测模式进行解析，根据属性预测模式选取对应的属性预测值；将属性预测值与属性预测残差信息相加，得到待解码点云的属性信息。属性信息可以包括属性坐标。
其中，可以依据属性量化参数确定第二属性量化步长，第二属性量化步长QS4可以为：
Figure PCTCN2022097635-appb-000008
其中，QP2可以表示属性量化参数。
进行反量化处理得到的属性预测残差信息RQtRes2可以为：
RQtRes2＝QtRes2·QS4
其中，QtRes2可以表示量化属性预测残差信息。
需要说明的是，本实施例作为与图3所示的实施例中对应的解码侧的实施方式，其具体的实施方式可以参见图3所示的实施例的相关说明，为了避免重复说明，本实施例不再赘述，且还可以达到相同有益效果。
需要说明的是，本申请实施例提供的点云编码处理方法，执行主体可以为点云编码处理装置，或者，该点云编码处理装置中的用于执行点云编码处理的方法的控制模块。本申请实施例中以点云编码处理装置执行点云编码处理的方法为例，说明本申请实施例提供的点云编码处理装置。
请参见图7，图7是本申请实施例提供的一种点云编码处理装置的结构图之一，如图7所示，点云编码处理装置300包括：
第一编码模块301，用于基于待编码点云的几何信息进行预测编码，得到几何预测残差信息；
第一量化模块302，用于依据几何量化参数对所述几何预测残差信息进行量化处理，得到量化几何预测残差信息；
第二编码模块303，用于基于所述量化几何预测残差信息进行熵编码，得到几何码流。
可选的，所述第一量化模块302具体用于：
确定几何量化控制参数是否指示启用量化处理；
在所述几何量化控制参数指示启用量化处理的情况下，依据几何量化参数对所述几何预测残差信息进行量化处理。
可选的，所述第一量化模块302具体还用于：
在所述几何量化控制参数指示不启用量化处理的情况下，依据所述几何预测残差信息进行熵编码，得到几何码流。
可选的，所述第一编码模块301具体用于：
基于待编码点云对应的节点标识将所述待编码点云划分为第一子待编码点云和第二子待编码点云；
在几何编码控制参数指示第一编码模式的情况下，对所述第一子待编码点云的几何信息进行预测编码；
在所述几何编码控制参数指示第二编码模式的情况下，对所述第二子待编码点云的几何信息进行预测编码。
可选的，如图8所示，所述第二编码模块303具体包括：
第一确定单元3031，用于基于所述量化几何预测残差信息确定至少两个候选几何预测残差信息；
第一获取单元3032，用于获取所述至少两个候选几何预测残差信息对应的率失真代价；
第二确定单元3033，用于依据所述至少两个候选几何预测残差信息对应的率失真代价确定目标量化几何预测残差信息；
第一编码单元3034，用于基于所述目标量化几何预测残差信息进行熵编码。
可选的，所述目标量化几何预测残差信息为所述至少两个候选几何预测残差信息中率失真代价最小的候选几何预测残差信息。
可选的，所述候选几何预测残差信息对应的率失真代价基于几何失真值及第一预测残差码率确定，所述几何失真值用于表征所述候选几何预测残差信息对应的几何失真，所述第一预测残差码率用于表征编码所述候选几何预测残差信息预计的比特值。
可选的，所述至少两个候选几何预测残差信息中包括与所述量化几何预测残差信息相关的候选几何预测残差信息，及与所述量化几何预测残差信息不相关的候选几何预测残差信息；
所述第一编码单元3034具体用于：
在所述目标量化几何预测残差信息为与所述量化几何预测残差信息相关的候选几何预测残差信息的情况下，基于所述目标量化几何预测残差信息对应的标识和所述目标量化几何预测残差信息进行熵编码；
在所述目标量化几何预测残差信息为与所述量化几何预测残差信息不相关的候选几何预测残差信息的情况下，基于所述目标量化几何预测残差信息对应的标识进行熵编码。
可选的，所述第一编码模块301具体用于：
依据预先设置的第一量化步长获取待编码点云对应的量化点云；
对所述量化点云进行去重处理；
将去重处理后得到的量化点云对应的待编码点云的几何信息进行预测编码。
可选的，所述第一量化模块302具体用于：
依据几何量化参数确定第一几何量化步长；
基于所述第一几何量化步长及第一预设几何偏移值对所述几何预测残差信息进行量化处理。
可选的，所述几何预测残差信息包括三个维度的子几何预测残差信息；
所述几何量化参数包括与所述三个维度的子几何预测残差信息分别对应的三个子几何量化参数。
可选的，如图9所示，所述装置300还包括：
第三编码模块304，用于对所述待编码点云的属性信息进行预测编码，得到属性预测残差信息；
第二量化模块305，用于依据属性量化参数对所述属性预测残差信息进行量化处理，得到量化属性预测残差信息；
第四编码模块306，用于基于所述量化属性预测残差信息进行熵编码，得到属性码流。
可选的，如图10所示，所述第四编码模块306具体包括：
第三确定单元3061，用于基于所述量化属性预测残差信息确定至少两个候选属性预测残差信息；
第二获取单元3062，用于获取所述至少两个候选属性预测残差信息对应的率失真代价；
第四确定单元3063，用于依据所述至少两个候选属性预测残差信息对应的率失真代价确定目标量化属性预测残差信息；
第二编码单元3064，用于基于所述目标量化属性预测残差信息进行熵编码。
可选的，所述目标量化属性预测残差信息为所述至少两个候选属性预测残差信息中率失真代价最小的候选属性预测残差信息。
可选的，所述候选属性预测残差信息对应的率失真代价基于属性失真值及第二预测残差码率确定，所述属性失真值用于表征所述候选属性预测残差信息对应的属性失真，所述第二预测残差码率用于表征编码所述候选属性预测残差信息预计的比特值。
可选的，所述至少两个候选属性预测残差信息中包括与所述量化属性预测残差信息相关的候选属性预测残差信息，及与所述量化属性预测残差信息不相关的候选属性预测残差信息；
所述第二编码单元3064具体用于：
在所述目标量化属性预测残差信息为与所述量化属性预测残差信息相关的候选属性预测残差信息的情况下，基于所述目标量化几何预测残差信息对应的标识和所述目标量化属性预测残差信息进行熵编码；
在所述目标量化属性预测残差信息为与所述量化属性预测残差信息不相关的候选属性预测残差信息的情况下，基于所述目标量化属性预测残差信息对应的标识进行熵编码。
本申请实施例中的点云编码处理装置300能够提高点云的几何码流的速率控制效果。
本申请实施例中的点云编码处理装置可以是装置，具有操作系统的装置或电子设备，也可以是终端中的部件、集成电路、或芯片。该装置或电子设 备可以是移动终端，也可以为非移动终端。示例性的，移动终端可以包括但不限于上述所列举的终端的类型，非移动终端可以为服务器、网络附属存储器(Network Attached Storage，NAS)、个人计算机(personal computer，PC)、电视机(television，TV)、柜员机或者自助机等，本申请实施例不作具体限定。
本申请实施例提供的点云编码处理装置能够实现图3的方法实施例实现的各个过程，并达到相同的技术效果，为避免重复，这里不再赘述。
需要说明的是，本申请实施例提供的点云解码处理方法，执行主体可以为点云解码处理装置，或者，该点云解码处理装置中的用于执行点云解码处理的方法的控制模块。本申请实施例中以点云解码处理装置执行点云解码处理的方法为例，说明本申请实施例提供的点云解码处理装置。
请参见图11，图11是本申请实施例提供的一种点云解码处理装置的结构图之一，如图11所示，点云解码处理装置400包括：
第一解码模块401，用于对几何码流进行熵解码，得到量化几何预测残差信息；
第一反量化模块402，用于依据几何量化参数对所述量化几何预测残差信息进行反量化处理，得到几何预测残差信息；
第二解码模块403，用于基于所述几何预测残差信息进行预测解码，得到待解码点云的几何信息。
可选的，所述第一解码模块401具体用于：
确定几何量化控制参数是否指示启用量化处理；
在所述几何量化控制参数指示启用量化处理的情况下，对所述几何码流进行熵解码，得到量化几何预测残差信息。
可选的，所述第一解码模块401具体还用于：
在所述几何量化控制参数指示不启用量化处理的情况下，对所述几何码流进行熵解码，得到几何预测残差信息。
可选的，所述第一反量化模块402具体用于：
依据几何量化参数确定第二几何量化步长；
基于所述第二几何量化步长及第二预设几何偏移值对所述量化几何预测 残差信息进行反量化处理。
可选的，所述几何预测残差信息包括三个维度的子几何预测残差信息；
所述几何量化参数包括与所述三个维度的子几何预测残差信息分别对应的三个子几何量化参数。
可选的，如图12所示，所述装置400还包括：
第三解码模块404，用于对属性码流进行熵解码，得到量化属性预测残差信息；
第二反量化模块405，用于依据属性量化参数对所述量化属性预测残差信息进行反量化处理，得到属性预测残差信息；
第四解码模块406，用于基于所述属性预测残差信息进行预测解码，得到所述待解码点云的属性信息。
本申请实施例中的点云解码处理装置400能够提高点云的几何码流的速率控制效果。
本申请实施例中的点云解码处理装置可以是装置，具有操作系统的装置或电子设备，也可以是终端中的部件、集成电路、或芯片。该装置或电子设备可以是移动终端，也可以为非移动终端。示例性的，移动终端可以包括但不限于上述所列举的终端的类型，非移动终端可以为服务器、网络附属存储器(Network Attached Storage，NAS)、个人计算机(personal computer，PC)、电视机(television，TV)、柜员机或者自助机等，本申请实施例不作具体限定。
本申请实施例提供的点云解码处理装置能够实现图6的方法实施例实现的各个过程，并达到相同的技术效果，为避免重复，这里不再赘述。
可选的，如图13所示，本申请实施例还提供一种通信设备500，包括处理器501，存储器502，存储在存储器502上并可在所述处理器501上运行的程序或指令，例如，该通信设备500为终端时，该程序或指令被处理器501执行时实现上述点云编码处理方法实施例的各个过程，且能达到相同的技术效果；或者，该程序或指令被处理器501执行时实现上述点云解码处理方法实施例的各个过程，且能达到相同的技术效果，为避免重复，这里不再赘述。
本申请实施例还提供一种终端，包括处理器和通信接口，该终端实施例 是与上述点云编码处理方法实施例对应的，或者，该终端实施例是与上述点云解码处理方法实施例对应的，上述方法实施例的各个实施过程和实现方式均可适用于该终端实施例中，且能达到相同的技术效果。具体地，图14为实现本申请实施例的一种终端的硬件结构示意图。
该终端600包括但不限于：射频单元601、网络模块602、音频输出单元603、输入单元604、传感器605、显示单元606、用户输入单元607、接口单元608、存储器609、以及处理器610等中的至少部分部件。
本领域技术人员可以理解终端600还可以包括给各个部件供电的电源(比如电池)，电源可以通过电源管理系统与处理器610逻辑相连，从而通过电源管理系统实现管理充电、放电、以及功耗管理等功能。图6中示出的终端结构并不构成对终端的限定，终端可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置，在此不再赘述。
应理解的是，本申请实施例中，输入单元604可以包括图形处理器(Graphics Processing Unit，GPU)6041和麦克风6042，图形处理器6041对在视频捕获模式或图像捕获模式中由图像捕获装置(如摄像头)获得的静态图片或视频的图像数据进行处理。显示单元606可包括显示面板6061，可以采用液晶显示器、有机发光二极管等形式来配置显示面板6061。用户输入单元607包括触控面板6071以及其他输入设备6072。触控面板6071，也称为触摸屏。触控面板6071可包括触摸检测装置和触摸控制器两个部分。其他输入设备6072可以包括但不限于物理键盘、功能键(比如音量控制按键、开关按键等)、轨迹球、鼠标、操作杆，在此不再赘述。
本申请实施例中，射频单元601将来自网络侧设备的下行数据接收后，给处理器610处理；另外，将上行的数据发送给网络侧设备。通常，射频单元601包括但不限于天线、至少一个放大器、收发信机、耦合器、低噪声放大器、双工器等。
存储器609可用于存储软件程序或指令以及各种数据。存储器609可主要包括存储程序或指令区和存储数据区，其中，存储程序或指令区可存储操作系统、至少一个功能所需的应用程序或指令(比如声音播放功能、图像播放功能等)等。此外，存储器609可以包括高速随机存取存储器，还可以包 括非易失性存储器，其中，非易失性存储器可以是只读存储器(Read-Only Memory，ROM)、可编程只读存储器(Programmable ROM，PROM)、可擦除可编程只读存储器(Erasable PROM，EPROM)、电可擦除可编程只读存储器(Electrically EPROM，EEPROM)或闪存。例如至少一个磁盘存储器件、闪存器件、或其他非易失性固态存储器件。
处理器610可包括一个或多个处理单元；可选的，处理器610可集成应用处理器和调制解调处理器，其中，应用处理器主要处理操作系统、用户界面和应用程序或指令等，调制解调处理器主要处理无线通信，如基带处理器。可以理解的是，上述调制解调处理器也可以不集成到处理器610中。
其中，所述终端是用于执行点云编码处理方法的情况下：
所述处理器或者所述通信接口用于：基于待编码点云的几何信息进行预测编码，得到几何预测残差信息；依据几何量化参数对所述几何预测残差信息进行量化处理，得到量化几何预测残差信息；基于所述量化几何预测残差信息进行熵编码，得到几何码流。
可选的，处理器610还用于：
确定几何量化控制参数是否指示启用量化处理；
在所述几何量化控制参数指示启用量化处理的情况下，依据几何量化参数对所述几何预测残差信息进行量化处理。
可选的，处理器610还用于：
在所述几何量化控制参数指示不启用量化处理的情况下，依据所述几何预测残差信息进行熵编码，得到几何码流。
可选的，处理器610还用于：
基于待编码点云对应的节点标识将所述待编码点云划分为第一子待编码点云和第二子待编码点云；
在几何编码控制参数指示第一编码模式的情况下，对所述第一子待编码点云的几何信息进行预测编码；
在所述几何编码控制参数指示第二编码模式的情况下，对所述第二子待编码点云的几何信息进行预测编码。
可选的，处理器610还用于：
基于所述量化几何预测残差信息确定至少两个候选几何预测残差信息；
获取所述至少两个候选几何预测残差信息对应的率失真代价；
依据所述至少两个候选几何预测残差信息对应的率失真代价确定目标量化几何预测残差信息；
基于所述目标量化几何预测残差信息进行熵编码。
可选的，所述目标量化几何预测残差信息为所述至少两个候选几何预测残差信息中率失真代价最小的候选几何预测残差信息。
可选的，所述候选几何预测残差信息对应的率失真代价基于几何失真值及第一预测残差码率确定，所述几何失真值用于表征所述候选几何预测残差信息对应的几何失真，所述第一预测残差码率用于表征编码所述候选几何预测残差信息预计的比特值。
可选的，所述至少两个候选几何预测残差信息中包括与所述量化几何预测残差信息相关的候选几何预测残差信息，及与所述量化几何预测残差信息不相关的候选几何预测残差信息；
处理器610还用于：
在所述目标量化几何预测残差信息为与所述量化几何预测残差信息相关的候选几何预测残差信息的情况下，基于所述目标量化几何预测残差信息对应的标识和所述目标量化几何预测残差信息进行熵编码；
在所述目标量化几何预测残差信息为与所述量化几何预测残差信息不相关的候选几何预测残差信息的情况下，基于所述目标量化几何预测残差信息对应的标识进行熵编码。
可选的，处理器610还用于：
依据预先设置的第一量化步长获取待编码点云对应的量化点云；
对所述量化点云进行去重处理；
将去重处理后得到的量化点云对应的待编码点云的几何信息进行预测编码。
可选的，处理器610还用于：
依据几何量化参数确定第一几何量化步长；
基于所述第一几何量化步长及第一预设几何偏移值对所述几何预测残差 信息进行量化处理。
可选的，所述几何预测残差信息包括三个维度的子几何预测残差信息；
所述几何量化参数包括与所述三个维度的子几何预测残差信息分别对应的三个子几何量化参数。
可选的，处理器610还用于：
对所述待编码点云的属性信息进行预测编码，得到属性预测残差信息；
依据属性量化参数对所述属性预测残差信息进行量化处理，得到量化属性预测残差信息；
基于所述量化属性预测残差信息进行熵编码，得到属性码流。
可选的，处理器610还用于：
基于所述量化属性预测残差信息确定至少两个候选属性预测残差信息；
获取所述至少两个候选属性预测残差信息对应的率失真代价；
依据所述至少两个候选属性预测残差信息对应的率失真代价确定目标量化属性预测残差信息；
基于所述目标量化属性预测残差信息进行熵编码。
可选的，所述目标量化属性预测残差信息为所述至少两个候选属性预测残差信息中率失真代价最小的候选属性预测残差信息。
可选的，所述候选属性预测残差信息对应的率失真代价基于属性失真值及第二预测残差码率确定，所述属性失真值用于表征所述候选属性预测残差信息对应的属性失真，所述第二预测残差码率用于表征编码所述候选属性预测残差信息预计的比特值。
可选的，关的候选属性预测残差信息，及与所述量化属性预测残差信息不相关的候选属性预测残差信息；
处理器610还用于：
在所述目标量化属性预测残差信息为与所述量化属性预测残差信息相关的候选属性预测残差信息的情况下，基于所述目标量化几何预测残差信息对应的标识和所述目标量化属性预测残差信息进行熵编码；
在所述目标量化属性预测残差信息为与所述量化属性预测残差信息不相关的候选属性预测残差信息的情况下，基于所述目标量化属性预测残差信息 对应的标识进行熵编码。
本申请实施例中的终端能够提高点云的几何码流的速率控制效果。
具体地，本申请实施例的终端还包括：存储在存储器609上并可在处理器610上运行的指令或程序，处理器610调用存储器609中的指令或程序执行图7所示各模块执行的方法，并达到相同的技术效果，为避免重复，故不在此赘述。
其中，所述终端是用于执行点云解码处理方法的情况下：
所述处理器或者所述通信接口用于：对几何码流进行熵解码，得到量化几何预测残差信息；依据几何量化参数对所述量化几何预测残差信息进行反量化处理，得到几何预测残差信息；基于所述几何预测残差信息进行预测解码，得到待解码点云的几何信息。
可选的，处理器610还用于：
确定几何量化控制参数是否指示启用量化处理；
在所述几何量化控制参数指示启用量化处理的情况下，对所述几何码流进行熵解码，得到量化几何预测残差信息。
可选的，处理器610还用于：
在所述几何量化控制参数指示不启用量化处理的情况下，对所述几何码流进行熵解码，得到几何预测残差信息。
可选的，处理器610还用于：
依据几何量化参数确定第二几何量化步长；
基于所述第二几何量化步长及第二预设几何偏移值对所述量化几何预测残差信息进行反量化处理。
可选的，所述几何预测残差信息包括三个维度的子几何预测残差信息；
所述几何量化参数包括与所述三个维度的子几何预测残差信息分别对应的三个子几何量化参数。
可选的，处理器610还用于：
对属性码流进行熵解码，得到量化属性预测残差信息；
依据属性量化参数对所述量化属性预测残差信息进行反量化处理，得到属性预测残差信息；
基于所述属性预测残差信息进行预测解码，得到所述待解码点云的属性信息。
本申请实施例中的终端能够提高点云的几何码流的速率控制效果。
具体地，本申请实施例的终端还包括：存储在存储器609上并可在处理器610上运行的指令或程序，处理器610调用存储器609中的指令或程序执行图11所示各模块执行的方法，并达到相同的技术效果，为避免重复，故不在此赘述。
本申请实施例还提供一种可读存储介质，所述可读存储介质上存储有程序或指令，该程序或指令被处理器执行时实现上述点云编码处理方法实施例的各个过程，或者，该程序或指令被处理器执行时实现上述点云解码处理方法实施例的各个过程，且能达到相同的技术效果，为避免重复，这里不再赘述。
其中，所述处理器为上述实施例中所述的终端中的处理器。所述可读存储介质，包括计算机可读存储介质，如计算机只读存储器(Read-Only Memory，ROM)、随机存取存储器(Random Access Memory，RAM)、磁碟或者光盘等。
本申请实施例另提供了一种芯片，所述芯片包括处理器和通信接口，所述通信接口和所述处理器耦合，所述处理器用于运行程序或指令，实现上述点云编码处理方法实施例的各个过程，或者，实现上述点云解码处理方法实施例的各个过程，且能达到相同的技术效果，为避免重复，这里不再赘述。
应理解，本申请实施例提到的芯片还可以称为系统级芯片，系统芯片，芯片系统或片上系统芯片等。
需要说明的是，在本文中，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者装置不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者装置所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括该要素的过程、方法、物品或者装置中还存在另外的相同要素。此外，需要指出的是，本申请实施方式中的方法和装置的范围不限按示出或讨论的顺序来执行功能，还 可包括根据所涉及的功能按基本同时的方式或按相反的顺序来执行功能，例如，可以按不同于所描述的次序来执行所描述的方法，并且还可以添加、省去、或组合各种步骤。另外，参照某些示例所描述的特征可在其他示例中被组合。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分可以以计算机程序产品的形式体现出来，该计算机程序产品存储在一个存储介质(如ROM/RAM、磁碟、光盘)中，包括若干指令用以使得一台终端(可以是手机，计算机，服务器，空调器，或者网络设备等)执行本申请各个实施例所述的方法。
上面结合附图对本申请的实施例进行了描述，但是本申请并不局限于上述的具体实施方式，上述的具体实施方式仅仅是示意性的，而不是限制性的，本领域的普通技术人员在本申请的启示下，在不脱离本申请宗旨和权利要求所保护的范围情况下，还可做出很多形式，均属于本申请的保护之内。

Claims (33)

  1. 一种点云编码处理方法，包括：
    基于待编码点云的几何信息进行预测编码，得到几何预测残差信息；
    依据几何量化参数对所述几何预测残差信息进行量化处理，得到量化几何预测残差信息；
    基于所述量化几何预测残差信息进行熵编码，得到几何码流。
  2. 根据权利要求1所述的方法，其中，所述依据几何量化参数对所述几何预测残差信息进行量化处理，包括：
    确定几何量化控制参数是否指示启用量化处理；
    在所述几何量化控制参数指示启用量化处理的情况下，依据几何量化参数对所述几何预测残差信息进行量化处理。
  3. 根据权利要求2所述的方法，其中，所述确定几何量化控制参数是否指示启用量化处理之后，所述方法还包括：
    在所述几何量化控制参数指示不启用量化处理的情况下，依据所述几何预测残差信息进行熵编码，得到几何码流。
  4. 根据权利要求1所述的方法，其中，所述对待编码点云的几何信息进行预测编码，包括：
    基于待编码点云对应的节点标识将所述待编码点云划分为第一子待编码点云和第二子待编码点云；
    在几何编码控制参数指示第一编码模式的情况下，对所述第一子待编码点云的几何信息进行预测编码；
    在所述几何编码控制参数指示第二编码模式的情况下，对所述第二子待编码点云的几何信息进行预测编码。
  5. 根据权利要求1所述的方法，其中，所述基于所述量化几何预测残差信息进行熵编码，包括：
    基于所述量化几何预测残差信息确定至少两个候选几何预测残差信息；
    获取所述至少两个候选几何预测残差信息对应的率失真代价；
    依据所述至少两个候选几何预测残差信息对应的率失真代价确定目标量化几何预测残差信息；
    基于所述目标量化几何预测残差信息进行熵编码。
  6. 根据权利要求5所述的方法，其中，所述目标量化几何预测残差信息为所述至少两个候选几何预测残差信息中率失真代价最小的候选几何预测残差信息。
  7. 根据权利要求5所述的方法，其中，所述候选几何预测残差信息对应的率失真代价基于几何失真值及第一预测残差码率确定，所述几何失真值用于表征所述候选几何预测残差信息对应的几何失真，所述第一预测残差码率用于表征编码所述候选几何预测残差信息预计的比特值。
  8. 根据权利要求5所述的方法，其中，所述至少两个候选几何预测残差信息中包括与所述量化几何预测残差信息相关的候选几何预测残差信息，及与所述量化几何预测残差信息不相关的候选几何预测残差信息；
    所述基于所述目标量化几何预测残差信息进行熵编码，包括：
    在所述目标量化几何预测残差信息为与所述量化几何预测残差信息相关的候选几何预测残差信息的情况下，基于所述目标量化几何预测残差信息对应的标识和所述目标量化几何预测残差信息进行熵编码；
    在所述目标量化几何预测残差信息为与所述量化几何预测残差信息不相关的候选几何预测残差信息的情况下，基于所述目标量化几何预测残差信息对应的标识进行熵编码。
  9. 根据权利要求1所述的方法，其中，所述对待编码点云的几何信息进行预测编码，包括：
    依据预先设置的第一量化步长获取待编码点云对应的量化点云；
    对所述量化点云进行去重处理；
    将去重处理后得到的量化点云对应的待编码点云的几何信息进行预测编 码。
  10. 根据权利要求1所述的方法，其中，所述依据几何量化参数对所述几何预测残差信息进行量化处理，包括：
    依据几何量化参数确定第一几何量化步长；
    基于所述第一几何量化步长及第一预设几何偏移值对所述几何预测残差信息进行量化处理。
  11. 根据权利要求1所述的方法，其中，所述几何预测残差信息包括三个维度的子几何预测残差信息；
    所述几何量化参数包括与所述三个维度的子几何预测残差信息分别对应的三个子几何量化参数。
  12. 根据权利要求1所述的方法，其中，所述方法还包括：
    对所述待编码点云的属性信息进行预测编码，得到属性预测残差信息；
    依据属性量化参数对所述属性预测残差信息进行量化处理，得到量化属性预测残差信息；
    基于所述量化属性预测残差信息进行熵编码，得到属性码流。
  13. 根据权利要求12所述的方法，其中，所述基于所述量化属性预测残差信息进行熵编码，包括：
    基于所述量化属性预测残差信息确定至少两个候选属性预测残差信息；
    获取所述至少两个候选属性预测残差信息对应的率失真代价；
    依据所述至少两个候选属性预测残差信息对应的率失真代价确定目标量化属性预测残差信息；
    基于所述目标量化属性预测残差信息进行熵编码。
  14. 根据权利要求13所述的方法，其中，所述目标量化属性预测残差信息为所述至少两个候选属性预测残差信息中率失真代价最小的候选属性预测残差信息。
  15. 根据权利要求13所述的方法，其中，所述候选属性预测残差信息对应的率失真代价基于属性失真值及第二预测残差码率确定，所述属性失真值 用于表征所述候选属性预测残差信息对应的属性失真，所述第二预测残差码率用于表征编码所述候选属性预测残差信息预计的比特值。
  16. 根据权利要求13所述的方法，其中，所述至少两个候选属性预测残差信息中包括与所述量化属性预测残差信息相关的候选属性预测残差信息，及与所述量化属性预测残差信息不相关的候选属性预测残差信息；
    所述基于所述目标量化属性预测残差信息进行熵编码，包括：
    在所述目标量化属性预测残差信息为与所述量化属性预测残差信息相关的候选属性预测残差信息的情况下，基于所述目标量化几何预测残差信息对应的标识和所述目标量化属性预测残差信息进行熵编码；
    在所述目标量化属性预测残差信息为与所述量化属性预测残差信息不相关的候选属性预测残差信息的情况下，基于所述目标量化属性预测残差信息对应的标识进行熵编码。
  17. 一种点云解码处理方法，包括：
    对几何码流进行熵解码，得到量化几何预测残差信息；
    依据几何量化参数对所述量化几何预测残差信息进行反量化处理，得到几何预测残差信息；
    基于所述几何预测残差信息进行预测解码，得到待解码点云的几何信息。
  18. 根据权利要求17所述的方法，其中，所述对所述几何码流进行熵解码，得到量化几何预测残差信息，包括：
    确定几何量化控制参数是否指示启用量化处理；
    在所述几何量化控制参数指示启用量化处理的情况下，对所述几何码流进行熵解码，得到量化几何预测残差信息。
  19. 根据权利要求18所述的方法，其中，所述确定几何量化控制参数是否指示启用量化处理之后，所述方法还包括：
    在所述几何量化控制参数指示不启用量化处理的情况下，对所述几何码流进行熵解码，得到几何预测残差信息。
  20. 根据权利要求17所述的方法，其中，所述依据几何量化参数对所述 量化几何预测残差信息进行反量化处理，包括：
    依据几何量化参数确定第二几何量化步长；
    基于所述第二几何量化步长及第二预设几何偏移值对所述量化几何预测残差信息进行反量化处理。
  21. 根据权利要求17所述的方法，其中，所述几何预测残差信息包括三个维度的子几何预测残差信息；
    所述几何量化参数包括与所述三个维度的子几何预测残差信息分别对应的三个子几何量化参数。
  22. 根据权利要求17所述的方法，其中，所述方法还包括：
    对属性码流进行熵解码，得到量化属性预测残差信息；
    依据属性量化参数对所述量化属性预测残差信息进行反量化处理，得到属性预测残差信息；
    基于所述属性预测残差信息进行预测解码，得到所述待解码点云的属性信息。
  23. 一种点云编码处理装置，包括：
    第一编码模块，用于基于待编码点云的几何信息进行预测编码，得到几何预测残差信息；
    第一量化模块，用于依据几何量化参数对所述几何预测残差信息进行量化处理，得到量化几何预测残差信息；
    第二编码模块，用于基于所述量化几何预测残差信息进行熵编码，得到几何码流。
  24. 根据权利要求23所述的装置，其中，所述第一量化模块具体用于：
    确定几何量化控制参数是否指示启用量化处理；
    在所述几何量化控制参数指示启用量化处理的情况下，依据几何量化参数对所述几何预测残差信息进行量化处理。
  25. 根据权利要求23所述的装置，其中，所述第二编码模块具体包括：
    第一确定单元，用于基于所述量化几何预测残差信息确定至少两个候选 几何预测残差信息；
    第一获取单元，用于获取所述至少两个候选几何预测残差信息对应的率失真代价；
    第二确定单元，用于依据所述至少两个候选几何预测残差信息对应的率失真代价确定目标量化几何预测残差信息；
    第一编码单元，用于基于所述目标量化几何预测残差信息进行熵编码。
  26. 根据权利要求23所述的装置，其中，所述装置还包括：
    第三编码模块，用于对所述待编码点云的属性信息进行预测编码，得到属性预测残差信息；
    第二量化模块，用于依据属性量化参数对所述属性预测残差信息进行量化处理，得到量化属性预测残差信息；
    第四编码模块，用于基于所述量化属性预测残差信息进行熵编码，得到属性码流。
  27. 根据权利要求26所述的装置，其中，所述第四编码模块具体包括：
    第三确定单元，用于基于所述量化属性预测残差信息确定至少两个候选属性预测残差信息；
    第二获取单元，用于获取所述至少两个候选属性预测残差信息对应的率失真代价；
    第四确定单元，用于依据所述至少两个候选属性预测残差信息对应的率失真代价确定目标量化属性预测残差信息；
    第二编码单元，用于基于所述目标量化属性预测残差信息进行熵编码。
  28. 一种点云解码处理装置，包括：
    第一解码模块，用于对几何码流进行熵解码，得到量化几何预测残差信息；
    第一反量化模块，用于依据几何量化参数对所述量化几何预测残差信息进行反量化处理，得到几何预测残差信息；
    第二解码模块，用于基于所述几何预测残差信息进行预测解码，得到待 解码点云的几何信息。
  29. 一种终端，包括处理器，存储器及存储在所述存储器上并可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现如权利要求1至16任一项所述的点云编码处理方法的步骤；或者，所述程序或指令被所述处理器执行时实现如权利要求17至22任一项所述的点云解码处理方法的步骤。
  30. 一种可读存储介质，所述可读存储介质上存储程序或指令，所述程序或指令被处理器执行时实现如权利要求1至16任一项所述的点云编码处理方法的步骤，或者，所述程序或指令被处理器执行时实现如权利要求17至22任一项所述的点云解码处理方法的步骤。
  31. 一种芯片，包括处理器和通信接口，其中，所述通信接口和所述处理器耦合，所述处理器用于运行程序或指令，实现如权利要求1至16任一项所述的点云编码处理方法的步骤，或者实现如权利要求17至22任一项所述的点云解码处理方法的步骤。
  32. 一种计算机程序产品，其中，所述程序产品被存储在非易失的存储介质中，所述程序产品被至少一个处理器执行以实现如权利要求1至16任一项所述的点云编码处理方法的步骤，或者实现如权利要求17至22任一项所述的点云解码处理方法的步骤。
  33. 一种通信设备，被配置为执行如权利要求1至16任一项所述的点云编码处理方法的步骤，或，执行如权利要求17至22任一项所述的点云解码处理方法的步骤。
PCT/CN2022/097635 2021-06-11 2022-06-08 点云编码处理方法、点云解码处理方法及相关设备 Ceased WO2022257971A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110656018.7 2021-06-11
CN202110656018.7A CN115474058B (zh) 2021-06-11 2021-06-11 点云编码处理方法、点云解码处理方法及相关设备

Publications (1)

Publication Number Publication Date
WO2022257971A1 true WO2022257971A1 (zh) 2022-12-15

Family

ID=84364387

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/097635 Ceased WO2022257971A1 (zh) 2021-06-11 2022-06-08 点云编码处理方法、点云解码处理方法及相关设备

Country Status (2)

Country Link
CN (1) CN115474058B (zh)
WO (1) WO2022257971A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20220108486A1 (en) * 2020-10-05 2022-04-07 Qualcomm Incorporated Scaling of quantization parameter values in geometry-based point cloud compression (g-pcc)
CN117579839A (zh) * 2024-01-15 2024-02-20 电子科技大学 一种基于率失真优化色彩空间转换矩阵的图像压缩方法
CN118283275A (zh) * 2022-12-30 2024-07-02 华为技术有限公司 编解码方法及电子设备
WO2025039120A1 (zh) * 2023-08-18 2025-02-27 Oppo广东移动通信有限公司 编解码方法、编解码器以及存储介质

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN121357324A (zh) * 2023-03-14 2026-01-16 维沃移动通信有限公司 点云编码处理方法、点云解码处理方法及相关设备
WO2024216435A1 (zh) * 2023-04-17 2024-10-24 Oppo广东移动通信有限公司 编解码方法及装置、编解码器、码流、设备、存储介质
WO2024216493A1 (zh) * 2023-04-18 2024-10-24 Oppo广东移动通信有限公司 点云编解码方法、编码器、解码器、码流及存储介质
WO2025007353A1 (zh) * 2023-07-06 2025-01-09 Oppo广东移动通信有限公司 编解码方法、码流、编码器、解码器以及存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111405281A (zh) * 2020-03-30 2020-07-10 北京大学深圳研究生院 一种点云属性信息的编码方法、解码方法、存储介质及终端设备
WO2020197228A1 (ko) * 2019-03-22 2020-10-01 엘지전자 주식회사 포인트 클라우드 데이터 송신 장치, 포인트 클라우드 데이터 송신 방법, 포인트 클라우드 데이터 수신 장치 및 포인트 클라우드 데이터 수신 방법
WO2021003173A1 (en) * 2019-07-02 2021-01-07 Tencent America LLC Method and apparatus for point cloud compression
CN112565764A (zh) * 2020-12-03 2021-03-26 西安电子科技大学 一种点云几何信息帧间编码及解码方法
CN112565757A (zh) * 2020-12-03 2021-03-26 西安电子科技大学 基于通道差异化的点云属性编码及解码方法、装置及系统

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3741118A1 (en) * 2018-01-19 2020-11-25 InterDigital VC Holdings, Inc. Processing a point cloud
CN112565794B (zh) * 2020-12-03 2022-10-04 西安电子科技大学 一种点云孤立点编码、解码方法及装置

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2020197228A1 (ko) * 2019-03-22 2020-10-01 엘지전자 주식회사 포인트 클라우드 데이터 송신 장치, 포인트 클라우드 데이터 송신 방법, 포인트 클라우드 데이터 수신 장치 및 포인트 클라우드 데이터 수신 방법
WO2021003173A1 (en) * 2019-07-02 2021-01-07 Tencent America LLC Method and apparatus for point cloud compression
CN111405281A (zh) * 2020-03-30 2020-07-10 北京大学深圳研究生院 一种点云属性信息的编码方法、解码方法、存储介质及终端设备
CN112565764A (zh) * 2020-12-03 2021-03-26 西安电子科技大学 一种点云几何信息帧间编码及解码方法
CN112565757A (zh) * 2020-12-03 2021-03-26 西安电子科技大学 基于通道差异化的点云属性编码及解码方法、装置及系统

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20220108486A1 (en) * 2020-10-05 2022-04-07 Qualcomm Incorporated Scaling of quantization parameter values in geometry-based point cloud compression (g-pcc)
US11869220B2 (en) * 2020-10-05 2024-01-09 Qualcomm Incorporated Scaling of quantization parameter values in geometry-based point cloud compression (G-PCC)
CN118283275A (zh) * 2022-12-30 2024-07-02 华为技术有限公司 编解码方法及电子设备
WO2024140683A1 (zh) * 2022-12-30 2024-07-04 华为技术有限公司 编解码方法及电子设备
WO2025039120A1 (zh) * 2023-08-18 2025-02-27 Oppo广东移动通信有限公司 编解码方法、编解码器以及存储介质
CN117579839A (zh) * 2024-01-15 2024-02-20 电子科技大学 一种基于率失真优化色彩空间转换矩阵的图像压缩方法
CN117579839B (zh) * 2024-01-15 2024-03-22 电子科技大学 一种基于率失真优化色彩空间转换矩阵的图像压缩方法

Also Published As

Publication number Publication date
CN115474058A (zh) 2022-12-13
CN115474058B (zh) 2025-06-24

Similar Documents

Publication Publication Date Title
WO2022257971A1 (zh) 点云编码处理方法、点云解码处理方法及相关设备
WO2022257970A1 (zh) 点云的几何信息编码处理方法、解码处理方法及相关设备
CN115474051B (zh) 点云编码方法、点云解码方法及终端
WO2022257969A1 (zh) 点云编码处理方法、解码处理方法及相关设备
WO2022257978A1 (zh) 点云编码方法、解码方法及装置
CN116233387B (zh) 点云编码、解码方法、装置及通信设备
WO2023025024A1 (zh) 点云属性编码方法、点云属性解码方法及终端
WO2022258009A1 (zh) 熵编码、解码方法及装置
WO2023098802A1 (zh) 点云属性编码方法、点云属性解码方法及终端
US20240121439A1 (en) Point cloud attribute information encoding method and apparatus, point cloud attribute information decoding method and apparatus, and related device
JP7660713B2 (ja) 点群コーディング処理方法、デコーディング処理方法及び装置
CN115474064B (zh) 编码、解码控制方法及装置
CN116233426A (zh) 属性量化、反量化方法、装置及设备
CN119815052B (zh) 编码方法、解码方法及相关设备
CN118678075B (zh) 点云编码处理方法、点云解码处理方法及相关设备
WO2023098803A1 (zh) 点云编码处理方法、点云解码处理方法及相关设备
CN120343259A (zh) 编解码方法及相关设备
WO2024245112A1 (zh) 点云编码处理方法、点云解码处理方法及相关设备
WO2024217301A1 (zh) 点云编码处理方法、点云解码处理方法及相关设备
CN120343269A (zh) 点云重建方法、装置及相关设备
CN120835158A (zh) Trisoup顶点优化方法、装置及设备

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22819568

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 22819568

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 22-04-2024)

122 Ep: pct application non-entry in european phase

Ref document number: 22819568

Country of ref document: EP

Kind code of ref document: A1