Point cloud encoding and decoding methods and codecs

Abstract

The present application discloses a point cloud encoding and decoding method and an encoding and decoding method, which relate to the technical field of encoding and decoding, and are helpful for improving encoding and decoding performance. The point cloud decoding method (including the point cloud encoding method or the point cloud decoding method) includes determining whether the spatial adjacent pixel blocks of the to-be-processed boundary pixel block of the occupancy map of the to-be-decoded point cloud are invalid pixel blocks, and determining whether the to-be-processed boundary pixel blocks The type of block; wherein, the invalid pixel block includes the pixel block whose value of the included pixels are not all 0 and does not belong to the target point cloud block patch occupancy map; or, the invalid pixel block includes the value of the included pixels are all 0 pixels. block, and the pixel blocks whose values ​​are not all 0 and do not belong to the target patch occupancy map; the target patch occupancy map is the patch occupancy map to which the boundary pixel block to be processed belongs; according to the type of the boundary pixel block to be processed, reconstruct The point cloud to be decoded.

Description

technical field [0001] The present application relates to the technical field of encoding and decoding, and in particular, to a point cloud (point cloud) encoding and decoding method and a codec. Background technique [0002] With the continuous development of 3D sensor (such as 3D scanner) technology, it is more and more convenient to collect point cloud data, and the scale of collected point cloud data is also getting larger and larger. How to effectively store massive point cloud data becomes A very urgent problem to be solved. High-quality compression, storage and transmission of point clouds becomes very important. [0003] In the process of reconstructing the point cloud, in order to eliminate the error of the reconstructed point cloud caused by the compression error of the point cloud occupancy map, it is usually necessary to perform special processing on the boundary pixel blocks of the reconstructed point cloud occupancy map, such as filling, Filtering, dilation o...

AI Technical Summary

Problems solved by technology

[0003] In the process of reconstructing the point cloud, in order to eliminate the error of the reconstructed point cloud caused by the compression error of the point cloud occupancy map, it is usually necessary to perform special processing on the boundary pixel blocks of the reconstructed point cloud occupancy map, such as filling, Filtering, dilation, erosion, etc.; in these processes, it is usually necessary to determine the processing method based on the local information of the boundary pixel blocks of the point cloud occupancy map, so it is necessary to classify these boundary image blocks, and to The accuracy of classification of boundary image blocks directly affects the encoding and decoding of point cloud occupancy maps, thereby affecting the effect of point cloud encoding and decoding

Images