Point cloud encoding and decoding method and apparatus based on two-dimensional regularized plane projection

Abstract

To disclose a point cloud encoding and decoding method and device based on a two-dimensional regularization plane projection.SOLUTION: An encoding method includes: acquiring original point cloud data; performing two-dimensional regularization plane projection on the original point cloud data to obtain a two-dimensional projection plane structure; obtaining a plurality of pieces of two-dimensional image information according to the two-dimensional projection plane structure; and encoding the plurality of pieces of two-dimensional image information to obtain code stream information. A point cloud in a three-dimensional space is projected to a corresponding two-dimensional regularization projection plane structure, and regularization correction is performed on the point cloud in a vertical direction and a horizontal direction, to obtain a strong correlation representation of the point cloud on the two-dimensional projection plane structure, so that the spatial correlation of the point cloud is better reflected. When the plurality of pieces of two-dimensional image information are subsequently encoded, the spatial correlation of the point cloud can be greatly utilized, and spatial redundancy is reduced, thereby further improving encoding efficiency of the point cloud.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] This application claims priority to Chinese Patent Application No. 202110172795.4, titled "Point Cloud Encoding and Decoding Method and Apparatus Based on 2D Regularized Planar Projection", filed with the China National Intellectual Property Administration on February 8, 2021, and incorporates it herein by reference in its entirety.

[0002] The present invention relates to the technical field of encoding and decoding, and specifically, to a point cloud encoding and decoding method and apparatus based on 2D regularized planar projection.

Background Art

[0003] With the improvement of hardware processing capabilities and the rapid development of computer vision, 3D point clouds have become a new generation of immersive multimedia following audio, images, and videos, and are widely applied to virtual reality, augmented reality, autonomous driving, environmental modeling, and the like. However, 3D point clouds usually have a relatively large amount of data that is not conducive to the transmission and storage of point cloud data. Therefore, it is of great significance to study efficient point cloud encoding and decoding technologies.

[0004] In the existing geometry-based point cloud compression (G-PCC) encoding framework, the geometry information and attribute information of the point cloud are encoded separately. Currently, G-PCC geometric encoding and decoding may be divided into octree-based geometric encoding and decoding and prediction tree-based geometric encoding and decoding.

[0005] Octree-based Geometric Encoding and Decoding: On the encoder side, first, the geometric information of the point cloud is preprocessed, which includes the coordinate transformation and voxelization process of the point cloud. Subsequently, in the breadth-first traversal order, tree splitting (octree / quadtree / binary tree) is continuously performed on the bounding box where the point cloud is located. Finally, the placeholder code of each node is encoded, and the number of points contained in each leaf node is encoded to generate a binary code stream. On the decoder side, first, the placeholder code of each node is continuously obtained by parsing in the breadth-first traversal order. Subsequently, tree splitting is continuously performed in sequence until a 1×1×1 unit cube is obtained through the splitting, and then the splitting stops. Finally, the number of points contained in each leaf node is obtained by parsing, and the finally reconstructed point cloud geometric information is obtained.

[0006] Prediction Tree-based Geometric Encoding and Decoding: On the encoder side, first, the original point cloud is sorted. Subsequently, a prediction tree structure is constructed. By classifying each point according to the laser scanner to which the point belongs, a prediction tree structure is constructed according to multiple different laser scanners. Subsequently, each node in the prediction tree is traversed, different prediction modes are selected to predict the geometric information of the node to obtain a prediction residual, and the prediction residual is quantized using quantization parameters. Finally, the prediction tree structure, quantization parameters, and prediction residuals of the geometric information of the nodes are encoded to generate a binary code stream. On the decoder side, first, the code stream is analyzed. Next, the prediction tree structure is reconstructed, and then, based on the prediction residual and quantization parameters of the geometric information of each node obtained by parsing, the prediction residual is dequantized, and finally, the reconstructed geometric information of each node is restored. That is, the reconstruction of the point cloud geometric information is completed.

[0007] However, due to the relatively strong spatial sparsity of the point cloud, in the point cloud encoding technology using the octree structure, this structure leads to a relatively large proportion of empty nodes being obtained by division, and it cannot fully reflect the spatial correlation of the point cloud, which is not conducive to point cloud prediction and entropy encoding. In the prediction tree-based point cloud encoding and decoding technology, a tree structure is constructed using some parameters of the lidar device, and this tree structure is used for the prediction encoding based on it. However, this tree structure does not fully reflect the spatial correlation of the point cloud, which is not conducive to point cloud prediction and entropy encoding. Therefore, both of the above two point cloud encoding and decoding technologies have the problem of insufficient high encoding efficiency.

Summary of the Invention

[0008] To solve the above problems in the prior art, the present invention provides a point cloud encoding and decoding method and device based on two-dimensional regularization plane projection. The technical problems solved by the present invention are realized by the following technical solutions.

[0009] A point cloud encoding method based on two-dimensional regularization plane projection is provided, and the method includes: acquiring original point cloud data; performing a two-dimensional regularization plane projection on the original point cloud data to obtain a two-dimensional projection plane structure; obtaining a plurality of two-dimensional image information according to the two-dimensional projection plane structure; encoding the plurality of two-dimensional image information to obtain code stream information.

[0010] In an embodiment of the present invention, performing a two-dimensional regularization plane projection on the original point cloud data to obtain a two-dimensional projection plane structure includes: initializing the two-dimensional projection plane structure; Determining a mapping relationship between the original point cloud data and the two-dimensional projection plane structure, and projecting the original point cloud data onto the two-dimensional projection plane structure.

[0011] In one embodiment of the present invention, the plurality of two-dimensional image information includes a geometry information map, and the geometry information map includes a placeholder information map, a depth information map, a projection residual information map, and a coordinate transformation error information map.

[0012] In one embodiment of the present invention, obtaining the code stream information by encoding the plurality of two-dimensional image information Encoding the placeholder information map, the depth information map, the projection residual information map, and the coordinate transformation error information map to obtain a placeholder information code stream, a depth information code stream, a projection residual information code stream, and a coordinate transformation error information code stream respectively; Obtaining a geometry information code stream according to the placeholder information code stream, the depth information code stream, the projection residual information code stream, and the coordinate transformation error information code stream.

[0013] In one embodiment of the present invention, after obtaining the geometry information code stream, the method further includes: Performing geometry reconstruction according to the geometry information code stream to obtain reconstructed point cloud geometry information; Encoding the attribute information of the original point cloud data based on the reconstructed point cloud geometry information to obtain an attribute information code stream.

[0014] In one embodiment of the present invention, the plurality of two-dimensional image information further includes an attribute information map.

[0015] In one embodiment of the present invention, obtaining the code stream information by encoding the plurality of two-dimensional image information further includes: Encoding an attribute information map to obtain an attribute information code stream.

[0016] Another embodiment of the present invention further provides a point cloud encoding device based on a two-dimensional regularized plane projection, and the device includes: A first data acquisition module configured to acquire original point cloud data; A projection module configured to perform a two-dimensional regularized plane projection on the original point cloud data to obtain a two-dimensional projection plane structure; A data processing module configured to obtain a plurality of two-dimensional image information according to the two-dimensional projection plane structure; An encoding module configured to encode the plurality of two-dimensional image information to obtain code stream information.

[0017] Yet another embodiment of the present invention further provides a point cloud decoding method based on a two-dimensional regularized plane projection, and the method includes: Acquiring code stream information, decoding the code stream information to obtain analyzed data; Reconstructing a plurality of two-dimensional image information according to the analyzed data; Obtaining a two-dimensional projection plane structure according to the plurality of two-dimensional image information; Reconstructing a point cloud using the two-dimensional projection plane structure.

[0018] Yet another embodiment of the present invention further provides a point cloud decoding device based on a two-dimensional regularized plane projection, and the device includes: A second data acquisition module configured to acquire code stream information, decoding the code stream information to obtain analyzed data; A first reconstruction module configured to reconstruct a plurality of two-dimensional image information according to the analyzed data; A second reconstruction module configured to obtain a two-dimensional projection plane structure according to the plurality of two-dimensional image information; A point cloud reconstruction module configured to reconstruct a point cloud using a two-dimensional projection plane structure.

[0019] The beneficial effects of the present invention are as follows: According to the present invention, by projecting a point cloud in a three-dimensional space onto a corresponding two-dimensional regularized projection plane structure and performing regularization correction on the point cloud in the vertical and horizontal directions to obtain a strong correlation representation of the point cloud on the two-dimensional projection plane structure, sparseness in the three-dimensional representation structure is avoided, and the spatial correlation of the point cloud is better reflected. When encoding a plurality of two-dimensional image information obtained for the two-dimensional regularized projection plane structure, the spatial correlation of the point cloud can be greatly utilized, spatial redundancy is reduced, and thereby the encoding efficiency of the point cloud is further improved.

[0020] Hereinafter, the present invention will be described in detail with reference to the accompanying drawings and embodiments.

Brief Description of the Drawings

[0021]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Embodiments for Carrying Out the Invention

[0022] Hereinafter, the present invention will be described in more detail with reference to specific embodiments, but the implementation of the present invention is not limited thereto.

[0023] Embodiment 1 FIG. 1 is a schematic diagram of a point cloud encoding method based on a two-dimensional regularized planar projection according to an embodiment of the present invention, and the method includes the following steps.

[0024] S1: Obtain original point cloud data.

[0025] Specifically, original point cloud data usually includes a group of three-dimensional space points, and each space point records its geometric position information and additional attribute information such as color, reflectivity, and normal. The geometric position information of the point cloud is generally represented based on the Cartesian coordinate system, that is, represented by the coordinates x, y, and z of the point. The original point cloud data may be obtained through a 3D scanning device such as a lidar, or alternatively, may be obtained based on publicly available datasets provided by various platforms. In this embodiment, it is assumed that the geometric position information of the obtained original point cloud data is represented based on the Cartesian coordinate system. Note that the representation method of the geometric position information of the original point cloud data is not limited to Cartesian coordinates.

[0026] S2: Perform a two-dimensional regularization plane projection on the original point cloud data to obtain a two-dimensional projection plane structure.

[0027] Specifically, in this embodiment, before performing a two-dimensional regularization plane projection on the original point cloud, in order to facilitate subsequent encoding, preprocessing such as voxelization processing is further performed on the original point cloud data.

[0028] First, the two-dimensional projection plane structure is initialized.

[0029] The initialization of the two-dimensional regularization projection plane structure of the point cloud requires the use of regularization parameters. The regularization parameters are usually precisely measured by the manufacturer and provided to the consumer as one of the required data. For example, the acquisition range of the lidar, the sampling angular resolution or the number of sampling points of the horizontal azimuth angle, the distance correction coefficient of each laser scanner, the offset information V o and H o of the laser scanner in the vertical and horizontal directions, and the offset information θ0 and α of the laser scanner along the pitch angle and the horizontal azimuth angle, etc.

[0030] Note that the regularization parameter is not limited to the parameters given above. A given calibration parameter of the lidar may be used as the regularization parameter, or when no calibration parameter of the lidar is given, the regularization parameter may be obtained by methods such as optimization of estimation and data fitting.

[0031] The two-dimensional regularized projection plane structure of the point cloud is a data structure including pixels of M rows and N columns, and the points in the three-dimensional point cloud correspond to the pixels in the projected data structure. Further, the pixel (i, j) in the data structure may be associated with the cylindrical coordinate components (θ, φ). For example, the pixel (i, j) corresponding to the cylindrical coordinates (r, θ, φ) is given by the following formula:

Number

[0032] Specifically, FIG. 2 is a schematic diagram of the correspondence between the cylindrical coordinates of points and the pixels in the two-dimensional projection plane according to an embodiment of the present invention.

[0033] Note that the correspondence of pixels here is not limited to cylindrical coordinates.

[0034] Also, the resolution of the two-dimensional regularized projection plane can be obtained using the regularization parameter. For example, assuming that the resolution of the two-dimensional regularized projection plane is M×N, M is initialized using the number of laser scanners in the regularization parameter, and the sampling angular resolution of the horizontal azimuth angle:

Number

Number

[0035] Furthermore, in order to project the original point cloud data onto the two-dimensional projection plane structure, the mapping relationship between the original point cloud data and the two-dimensional projection plane structure is determined.

[0036] In this part, by determining the position of the original point cloud within the two-dimensional projection plane structure for each point, the point cloud that was originally distributed disorderly in the Cartesian coordinate system is mapped onto a uniformly distributed two-dimensional regularized projection plane structure. Specifically, for each point in the original point cloud, the corresponding pixel within the two-dimensional projection plane structure is determined. For example, the pixel with the shortest spatial distance from the projection position of that point in the two-dimensional plane can be selected as the corresponding pixel for that point.

[0037] When the cylindrical coordinate system is used for the two-dimensional projection, the specific process for determining the pixel corresponding to the original point cloud is as follows.

[0038] a. The cylindrical coordinate component r of the current point of the original point cloud data is determined. Specifically, the following formula:

Equation

[0039] b. The search area of the current point within the two-dimensional projection plane structure is determined. Specifically, the entire two-dimensional projection plane structure may be directly selected as the search area. Also, in order to reduce the computational amount, the search area of the corresponding pixel within the two-dimensional projection plane structure can be determined by further using the pitch angle θ and azimuth angle φ of the cylindrical coordinate component of the current point, so as to reduce the search area.

[0040] c. After the exploration area is determined, for each pixel (i, j) within the exploration area, that is, the calibration parameters θ0, V of the i-th laser scanner of the lidar o , H o , and the regularization parameter α which is α, calculate the position (xl, yl, zl) of the current pixel in the Cartesian coordinate system. The specific calculation formula is as follows:

Equation

[0041] d. After obtaining the position (xl, yl, zl) of the current pixel in the Cartesian coordinate system, the spatial distance between this position and the current point (x, y, z) is calculated, and the error Err, that is,

Equation

[0042] If the error Err is smaller than the current minimum error minErr, update the minimum error minErr using the error Err, and update the i and j corresponding to the current pixel using the i and j corresponding to the current pixel. When the error Err is larger than the minimum error minErr, the above update process will not be performed.

[0043] e. After traversing all the pixels within the exploration area, the corresponding pixel (i, j) of the current point in the two-dimensional projection plane structure can be determined.

[0044] When the above processing is completed for all the points in the original point cloud, the two-dimensional regularization plane projection of the point cloud is completed. Specifically, FIG. 3 is a schematic diagram of the two-dimensional projection plane structure of the point cloud according to an embodiment of the present invention. Each point of the original point cloud data is mapped to the corresponding pixel within this structure.

[0045] In the two-dimensional regularization plane projection of the point cloud, multiple points in the point cloud may correspond to the same pixel in the two-dimensional projection plane structure. To avoid this situation, these spatial points can be selected to be projected onto different pixels during projection. For example, during the projection of a specific point, if the pixel corresponding to the point already has a corresponding point, the point is projected onto an empty pixel adjacent to the pixel. Further, when multiple points in the point cloud are projected onto the same pixel in the two-dimensional projection plane structure, during the encoding based on the two-dimensional projection plane structure, the number of corresponding points in each pixel should be additionally encoded, and according to this number of points, the information of each corresponding point in the pixel is encoded.

[0046] S3: Obtain a plurality of two-dimensional image information according to the two-dimensional projection plane structure.

[0047] In this embodiment, the above-mentioned plurality of two-dimensional image information may include a geometry information map. The geometry information map can be one or more of a placeholder information map, a depth information map, a projection residual information map, and a coordinate transformation error information map, or may be another geometry information map.

[0048] In this embodiment, specifically, a detailed description will be given by taking the above four geometry information maps as examples.

[0049] a. Placeholder information map The placeholder information map is used to identify whether each pixel in the two-dimensional regularized projection plane structure is occupied, that is, whether each pixel corresponds to a point in the point cloud. If each pixel is occupied, the pixel is referred to as non-empty, and if not, the pixel is referred to as empty. For example, 0 and 1 may be used for representation, where 1 indicates that the current pixel is occupied and 0 indicates that the current pixel is not occupied. In this way, the placeholder information map of the point cloud can be obtained according to the two-dimensional projection plane structure of the point cloud.

[0050] b. Depth information map The depth information map is used to represent the distance between the corresponding point of each occupied pixel in the two-dimensional regularized projection plane structure and the coordinate origin. For example, the cylindrical coordinate component r of the point corresponding to the pixel can be used as the depth of the pixel. Assuming that the Cartesian coordinates of the point corresponding to the pixel are (x, y, z), the cylindrical coordinate component r of the point, that is, the depth of the pixel, is given by the formula:

Equation

[0051] c. Projection residual information map The projection residual information map is used to represent the residual between the corresponding position and the actual projection position of each occupied pixel in the two-dimensional regularized projection plane structure. FIG. 4 is a schematic diagram of the projection residual according to an embodiment of the present invention.

[0052] Specifically, the projection residual of the pixel can be calculated as follows. Assuming that the current pixel is (i, j) and the Cartesian coordinates of the corresponding point of the current pixel are (x, y, z), the actual projection position of the point can be represented as (φ’, i’), which is given by the following formula:

Number

[0053] The corresponding position of the current pixel can be represented as (φ, i), which is given by the following equation:

Number

[0054] Therefore, the projection residual (Δφ, Δi) corresponding to the current pixel is given by the following equation:

Number

[0055] Based on the above calculations, each occupied pixel in the two-dimensional regularized projection plane has a projection residual, and as a result, a projection residual information map corresponding to the point cloud is obtained.

[0056] d. Coordinate transformation error information map The coordinate transformation error information map is used to represent the residual between the spatial position obtained through the back-projection of each occupied pixel in the two-dimensional regularized projection plane structure and the spatial position of the original point corresponding to that pixel.

[0057] For example, the coordinate transformation error of a pixel can be calculated as follows. Assume that the current pixel is (i, j) and the Cartesian coordinates of the corresponding point of the current pixel are (x, y, z). Then, using the regularization parameter and the following equation, the pixel can be inverse-transformed back to the Cartesian coordinate system to obtain the corresponding Cartesian coordinates (xl, yl, zl):

Number

[0058] After that, the coordinate transformation error (Δx, Δy, Δz) of the current pixel is given by the following equation:

Number

[0059] Based on the above calculation, each occupied pixel in the two-dimensional regularized projection plane structure has a coordinate transformation error, and as a result, a coordinate transformation error information map corresponding to the point cloud is obtained.

[0060] According to the present invention, by projecting the point cloud in the three-dimensional space onto the corresponding two-dimensional regularized projection plane structure and performing regularization correction on the point cloud in the vertical and horizontal directions to obtain a strong correlation representation of the point cloud on the two-dimensional projection plane structure, the spatial correlation of the point cloud is better reflected, thereby further improving the coding efficiency of the point cloud.

[0061] S4: Encoding the above plurality of two-dimensional image information to obtain code stream information.

[0062] FIG. 5 is an encoding framework diagram of point cloud geometry information according to an embodiment of the present invention.

[0063] In this embodiment, the placeholder information map, depth information map, projection residual information map, and coordinate transformation error information map obtained in step S3 are encoded to obtain a placeholder information code stream, a depth information code stream, a projection residual information code stream, and a coordinate transformation error information code stream, respectively.

[0064] Specifically, in this embodiment, the pixels in the placeholder information map, depth information map, projection residual information map, and coordinate transformation error information map are separately traversed in a specific scan order, such as a Z-shaped scan.

[0065] For the current pixel in the placeholder information map, the reconstructed placeholder information of the encoded and decoded pixels can be used for prediction. Specifically, various existing neighboring prediction techniques can be used. After obtaining the corresponding prediction residual, an existing entropy encoding technique can be used for encoding to obtain a placeholder information code stream.

[0066] For the current pixel in the depth information map, the reconstructed placeholder information map and the reconstructed depth information of the encoded and decoded pixels can be used for prediction. Specifically, prediction can be performed based on existing neighboring prediction techniques combined with the placeholder information of neighboring pixels, that is, predicting the depth information of the current pixel using only neighboring pixels with non-empty placeholder information. The predicted value can be calculated by a method such as weighted average. After obtaining the corresponding prediction residual, an existing entropy encoding technique can be used for encoding to obtain a depth information code stream.

[0067] For the current pixel in the projection residual information map, the reconstructed placeholder information map and depth information map, and the reconstructed projection residual information of the encoded and decoded pixels can be used for prediction. Specifically, prediction can be performed based on existing neighboring prediction techniques combined with the placeholder information and depth information of neighboring pixels, that is, predicting the projection residual information of the current pixel using only neighboring pixels with non-empty placeholder information and depth information close to the depth information of the current pixel. The predicted value can be calculated by a method such as weighted average. After obtaining the corresponding prediction residual, an existing entropy encoding technique can be used for encoding to obtain a prediction residual information code stream.

[0068] For the current pixel in the coordinate transformation error information map, the reconstructed placeholder information map, depth information map, and projection residual information map, as well as the reconstructed coordinate transformation error information of the encoded and decoded pixels, can be used for prediction. Specifically, prediction can be performed based on existing neighboring prediction techniques combined with the placeholder information, depth information, and projection residual information of neighboring pixels, that is, only neighboring pixels with non-empty placeholder information and depth information and projection residual information close to those of the current pixel are used to predict the coordinate transformation error information of the current pixel. The predicted value can be calculated by a method such as weighted average. After obtaining the corresponding prediction residual, an existing entropy encoding technique can be used for encoding to obtain a coordinate transformation error information code stream.

[0069] Alternatively, the prediction residual may be quantized and then encoded.

[0070] In this embodiment, when encoding a plurality of two-dimensional map information obtained by two-dimensional regularized planar projection, such as a placeholder information map, a depth information map, a projection residual information map, and a coordinate transformation error information map, strong correlations in the two-dimensional image are effectively used to perform prediction and entropy encoding for the current pixel, so that the spatial correlation of the point cloud can be greatly utilized, the spatial redundancy can be reduced, and thereby the encoding efficiency of the point cloud is further improved.

[0071] According to the placeholder information code stream, depth information code stream, projection residual information code stream, and coordinate transformation error information code stream, a geometry information code stream is obtained.

[0072] In this embodiment, after encoding all the geometry information maps, a geometry information code stream of the original point cloud data can be obtained.

[0073] In another embodiment of the present invention, for the aforementioned geometry information map, compression may alternatively be performed through image / video compression including, but not limited to, JPEG, JPEG2000, HEIF, H.264|AVC, and H.265|HEVC, etc.

[0074] It should also be noted that changes and adjustments to the projection residuals of the point cloud will affect the magnitude of its coordinate transformation error. Therefore, in order to obtain higher coding efficiency, the projection residuals and coordinate transformation errors of the point cloud may be adjusted rather than fixed. For example, in reversible coding, when the two-dimensional projection accuracy is relatively high, the projection residuals will be relatively small. Therefore, the projection residuals may be set to 0. Although this operation slightly increases the coordinate transformation error, if the coding efficiency of the coordinate transformation error is allowed to have a certain decrease or remain unchanged, the coding efficiency of the projection residuals can be significantly improved. Alternatively, when the accuracy of the two-dimensional projection is relatively low, the projection residuals will be relatively large. In this case, the projection residuals may be appropriately adjusted, and as a result, the coordinate transformation error will change accordingly. Subsequently, the adjusted projection residuals and coordinate transformation errors are coded to obtain higher coding efficiency. In irreversible coding, when the two-dimensional projection accuracy is relatively high, the projection residuals will be relatively small, and therefore, it may not be necessary to code the projection residuals. Alternatively, when the coordinate transformation error is relatively small, it may not be necessary to code the coordinate transformation error, thereby improving the coding efficiency. Instead, the projection residuals and coordinate transformation errors may be appropriately adjusted, and the adjusted projection residuals and adjusted coordinate transformation errors are coded to obtain higher coding efficiency.

[0075] Embodiment 2 Based on the completion of the coding of the geometry information in Embodiment 1, the coding of the attribute information can be further performed based on the reconstructed geometry information. FIG. 6 is a framework diagram for performing the coding of the attribute information based on the reconstructed geometry information according to an embodiment of the present invention.

[0076] First, perform geometry reconstruction according to the geometry information code stream obtained in Embodiment 1 to obtain reconstructed point cloud geometry information.

[0077] After that, encode the attribute information of the original point cloud data based on the reconstructed point cloud geometry information to obtain an attribute information code stream.

[0078] Specifically, the encoding of the attribute information is generally performed on the color information and reflectivity information of the spatial points. The attribute information of the original point cloud data can be encoded using existing techniques based on the geometric reconstruction information of the point cloud. For example, the color information in the attribute is first converted from the RGB color space to the YUV color space. Subsequently, the point cloud is recolored based on the reconstructed geometry information, and thus, the non-encoded attribute information corresponds to the reconstructed geometry information. After sorting the point cloud using the Morton code or Hilbert code, an interpolation prediction is performed on the points to be predicted using the reconstructed attribute values of the encoded points to obtain predicted attribute values, and then a difference is taken between the actual attribute values and the predicted attribute values to obtain prediction residuals. Finally, the prediction residuals are quantized and encoded to generate a binary code stream.

[0079] Embodiment 3 Based on Embodiment 1, by using a two-dimensional projection plane structure, an attribute information map can also be obtained simultaneously. Then, the geometry information map and the attribute information map are encoded simultaneously to obtain a geometry information code stream and an attribute information code stream. FIG. 7 is a framework diagram for simultaneously encoding point cloud geometry information and attribute information according to an embodiment of the present invention.

[0080] Specifically, for the encoding process of the geometry information map, reference can be made to Embodiment 1. The encoding process of the attribute information map is as follows.

[0081] First, pixels in the attribute information map are traversed in a specific scan order, such as through a zigzag scan. Subsequently, for the current pixel in the attribute information map, the reconstructed placeholder information map and the reconstructed attribute information of the encoded and decoded pixels can be used for prediction. Specifically, prediction can be performed based on existing neighboring prediction techniques combined with the placeholder information of neighboring pixels, that is, the attribute information of the current pixel is predicted using only neighboring pixels with non-empty placeholder information. The predicted value can be calculated by a method such as weighted average. After obtaining the corresponding prediction residual, an existing entropy encoding technique can be used for encoding to obtain an attribute information code stream.

[0082] In this embodiment, the geometry information map and the attribute information map are simultaneously acquired using a two-dimensional projection plane structure, and the geometry information and the attribute information are simultaneously encoded, thereby improving the encoding efficiency.

[0083] Embodiment 4 Based on Embodiments 1 to 3, this embodiment provides a point cloud encoding device based on two-dimensional regularization plane projection. FIG. 8 is a schematic configuration diagram of a point cloud encoding device based on two-dimensional regularization plane projection according to an embodiment of the present invention, and the device includes a first data acquisition module 11 configured to acquire original point cloud data, a projection module 12 configured to perform a two-dimensional regularization plane projection on the original point cloud data to obtain a two-dimensional projection plane structure, a data processing module 13 configured to obtain a plurality of two-dimensional image information according to the two-dimensional projection plane structure, and an encoding module 14 configured to encode the plurality of two-dimensional image information to obtain code stream information.

[0084] The encoding device provided in this embodiment can implement the encoding methods described in Embodiments 1 to 3, and the detailed process will not be described again here.

[0085] Embodiment 5 FIG. 9 is a schematic diagram of a point cloud decoding method based on two-dimensional regularization plane projection according to an embodiment of the present invention, and the method includes the following.

[0086] Step 1: Obtain code stream information, and decode the code stream information to obtain parsed data.

[0087] On the decoder side, compressed code stream information is obtained, and corresponding decoding is performed on the code stream information using corresponding existing entropy decoding techniques to obtain parsed data.

[0088] Step 2: Reconstruct a plurality of two-dimensional image information according to the parsed data.

[0089] In this embodiment, the plurality of two-dimensional image information includes a geometry information map, and the geometry information map includes a placeholder information map, a depth information map, a projection residual information map, and a coordinate transformation error information map.

[0090] Based on this, the parsed data mainly includes the prediction residual of the placeholder information, the prediction residual of the depth information, the prediction residual of the projection residual information, and the prediction residual of the coordinate transformation error information.

[0091] On the encoder side, the pixels in the placeholder information map, depth information map, projection residual information map, and coordinate transformation error information map are traversed separately in a specific scan order to encode the corresponding information. Therefore, the predicted residual information of the pixels obtained on the decoder side is also in this order. The decoder side can obtain the resolution of the two-dimensional map by using the regularization parameter. For details, reference can be made to the part of initializing the two-dimensional projection plane structure in S2 of Embodiment 1. Therefore, the decoder side can know the position of the pixel to be currently reconstructed in the two-dimensional map.

[0092] For the pixel to be currently reconstructed in the placeholder information map, the reconstructed placeholder information of the encoded and decoded pixels is used for prediction. The prediction method is consistent with that on the encoder side. Then, the placeholder information of the current pixel is reconstructed according to the obtained predicted value and the analyzed predicted residual.

[0093] For the pixel to be currently reconstructed in the depth information map, the reconstructed placeholder information map and the reconstructed depth information of the encoded and decoded pixels can be used for prediction. The prediction method is consistent with that on the encoder side, that is, the depth information of the current pixel is predicted only using neighboring pixels with non-empty placeholder information, and then the depth information of the current pixel is reconstructed according to the obtained predicted value and the analyzed predicted residual.

[0094] For the pixel to be currently reconstructed in the projection residual information map, the reconstructed placeholder information map, the reconstructed depth information map, and the reconstructed projection residual information of the encoded and decoded pixels can be used for prediction. The prediction method is consistent with that on the encoder side, that is, the projection residual information of the current pixel is predicted only using neighboring pixels with non-empty placeholder information and depth information close to the depth information of the current pixel, and then the projection residual information of the current pixel is reconstructed according to the obtained predicted value and the analyzed predicted residual.

[0095] For the pixels to be currently reconstructed in the coordinate transformation error information map, the reconstructed placeholder information map, the reconstructed depth information map, the reconstructed projection residual information map, and the reconstructed coordinate transformation error information of the encoded and decoded pixels can be used for prediction. The prediction method is consistent with that on the encoder side, that is, only neighboring pixels with non-empty placeholder information and depth information and projection residual information close to those of the current pixel are used to predict the coordinate transformation error information of the current pixel, and then the coordinate transformation error information of the current pixel is reconstructed according to the obtained predicted value and the analyzed prediction residual.

[0096] After each pixel in the placeholder information map, the depth information map, the projection residual information map, and the coordinate transformation error information map is reconstructed, a reconstructed placeholder information map, a reconstructed depth information map, a reconstructed projection residual information map, and a reconstructed coordinate transformation error information map can be obtained.

[0097] Also, the analyzed data can further include the prediction residual of the attribute information, and correspondingly, the attribute information map can also be reconstructed based on the information.

[0098] Step 3: Obtain a two-dimensional projection plane structure according to the above-mentioned plurality of two-dimensional image information.

[0099] The resolution of the two-dimensional projection plane structure is consistent with the resolution of the placeholder information map, the depth information map, the projection residual information map, and the coordinate transformation error information map, and since all the two-dimensional image information has been reconstructed, the placeholder information, the depth information, the projection residual information, and the coordinate transformation error information of each pixel in the two-dimensional projection plane structure can be known, and a reconstructed two-dimensional projection plane structure can be obtained.

[0100] Correspondingly, the reconstructed two-dimensional projection plane structure can further include the attribute information of the point cloud.

[0101] Step 4: Reconstruct the point cloud using the two-dimensional projection plane structure.

[0102] Specifically, by traversing the pixels in a specific scan order within the reconstructed two-dimensional projection plane structure, the placeholder information, depth information, projection residual information, and coordinate transformation error information of each pixel can be obtained. If the placeholder information of the current pixel (i, j) is non-empty, the spatial point (x, y, z) corresponding to the pixel, that is, the depth information of the current pixel which is the cylindrical coordinate component r of the corresponding point of the pixel, the projection residual information which is the residual (Δφ, Δi) between the corresponding position and the actual projection position of the pixel, and the coordinate transformation error information which is the residual (Δx, Δy, Δz) between the spatial position obtained by the back-projection of the pixel and the spatial position of the original point corresponding to the pixel, can be reconstructed as follows.

[0103] The corresponding position of the current pixel (i, j) can be represented as (φj, i), and then the actual projection position (φ’, i’) of the spatial point corresponding to the current pixel is as follows:

Equation

[0104] Using the regularization parameter and the following equation, the current pixel can be inverse-transformed back to the Cartesian coordinate system to obtain the corresponding Cartesian coordinates (xl, yl, zl):

Equation

[0105] The spatial point (x, y, z) corresponding to the current pixel is reconstructed using the following equation according to the spatial position (xl, yl, zl) obtained by the back-projection of the current pixel and the coordinate transformation error (Δx, Δy, Δz):

Equation

[0106] For each non-empty pixel in the two-dimensional projection structure, the corresponding spatial point can be reconstructed according to the above formula to obtain a reconstructed point cloud.

[0107] Note that when the point cloud is reconstructed on the decoder side, the reconstruction method can be adaptively selected according to the encoding method of the geometric information and attribute information of the point cloud at the encoder end to obtain the corresponding reconstructed point cloud.

[0108] Embodiment 6 Based on Embodiment 5, this embodiment provides a point cloud decoding device based on two-dimensional regularized plane projection. FIG. 10 is a schematic configuration diagram of a point cloud decoding device based on two-dimensional regularized plane projection according to an embodiment of the present invention. The device includes a second data acquisition module 21 configured to acquire code stream information and decode the code stream information to obtain analyzed data, a first reconstruction module 22 configured to reconstruct a plurality of two-dimensional image information according to the analyzed data, a second reconstruction module 23 configured to obtain a two-dimensional projection plane structure according to the plurality of two-dimensional image information, and a point cloud reconstruction module 24 configured to reconstruct a point cloud using the two-dimensional projection plane structure.

[0109] The decoding device provided in this embodiment can implement the decoding method of Embodiment 5, and the detailed process will not be described again here.

[0110] The above content is a detailed description of the present invention with reference to specific exemplary embodiments, and the specific implementation of the present invention should not be regarded as limited to these descriptions. Those skilled in the art to which the present invention pertains can make several simple deductions or substitutions without departing from the concept of the present invention, and all such deductions or substitutions should be regarded as being within the protection scope of the present invention.

Claims

1. Obtaining original point cloud data; Performing a two-dimensional regularization plane projection on the original point cloud data to obtain one two-dimensional projection plane structure; Obtaining a plurality of two-dimensional image information according to the two-dimensional projection plane structure, wherein the plurality of two-dimensional image information has a geometry information map; Encoding the plurality of two-dimensional image information to obtain code stream information; A point cloud encoding method based on a two-dimensional regularization plane projection having the above.

2. Performing a two-dimensional regularization plane projection on the original point cloud data to obtain a two-dimensional projection plane structure includes: Initializing the two-dimensional projection plane structure; Determining a mapping relationship between the original point cloud data and the two-dimensional projection plane structure, and projecting the original point cloud data onto the two-dimensional projection plane structure; The point cloud encoding method based on the two-dimensional regularization plane projection according to Claim 1, having the above.

3. The point cloud encoding method based on the two-dimensional regularization plane projection according to Claim 1, wherein the geometry information map has a placeholder information map.

4. Encoding the plurality of two-dimensional image information to obtain code stream information includes: Encoding the placeholder information map to obtain a placeholder information code stream; Obtaining a geometry information code stream according to the placeholder information code stream; The point cloud encoding method based on the two-dimensional regularization plane projection according to Claim 3, having the above.

5. After obtaining the geometry information code stream, the method further includes: Performing geometry reconstruction according to the geometry information code stream to obtain reconstructed point cloud geometry information; Encoding the attribute information of the original point cloud data based on the reconstructed point cloud geometry information to obtain an attribute information code stream; The point cloud encoding method based on the two-dimensional regularization plane projection according to Claim 4, having the above.

6. The point cloud encoding method based on the two-dimensional regularization plane projection according to Claim 3, wherein the plurality of two-dimensional image information further has an attribute information map.

7. Encoding the plurality of two-dimensional image information to obtain code stream information further includes: Encoding the attribute information map to obtain an attribute information code stream; The point cloud encoding method based on the two-dimensional regularization plane projection according to claim 6, which has...

8. A first data acquisition module (11) configured to acquire original point cloud data; A projection module (12) configured to perform a two-dimensional regularization plane projection on the original point cloud data to obtain one two-dimensional projection plane structure; A data processing module (13) configured to obtain a plurality of two-dimensional image information according to the two-dimensional projection plane structure, wherein the plurality of two-dimensional image information has a geometry information map; a processing module; An encoding module (14) configured to encode the plurality of two-dimensional image information to obtain code stream information; A point cloud encoding device based on the two-dimensional regularization plane projection, which has...

9. Obtaining code stream information, decoding the code stream information to obtain analyzed data; Reconstructing a plurality of two-dimensional image information according to the analyzed data; Obtaining one two-dimensional projection plane structure according to the plurality of two-dimensional image information, wherein the plurality of two-dimensional image information has a geometry information map; Reconstructing a point cloud using the two-dimensional projection plane structure; A point cloud decoding method based on the two-dimensional regularization plane projection, which has...

10. A second data acquisition module (21) configured to acquire code stream information and decode the code stream information to obtain analyzed data; A first reconstruction module (22) configured to reconstruct a plurality of two-dimensional image information according to the analyzed data; A second reconstruction module (23) configured to obtain one two-dimensional projection plane structure according to the plurality of two-dimensional image information, wherein the plurality of two-dimensional image information has a geometry information map; a second reconstruction module; A point cloud reconstruction module (24) configured to reconstruct a point cloud using the two-dimensional projection plane structure; A point cloud decoding device based on the two-dimensional regularization plane projection, which has...

11. A computer-readable storage medium storing computer-executable program instructions, wherein when the computer-executable program instructions are executed on a computer, the computer is enabled to execute the method according to any one of claims 1 to 7.

12. A computer-readable storage medium storing computer-executable program instructions, wherein when the computer-executable program instructions are executed on a computer, the computer is enabled to execute the method according to claim 9.

Images