Point cloud neighbor determination, point cloud prediction, point cloud coding and point cloud decoding method and device

Abstract

The invention discloses a point cloud neighbor determination, point cloud prediction, point cloud coding and point cloud decoding method and device. The method comprises the steps: determining a parameter p for comparing the distance; determining a plurality of neighbor candidate points of the current point; according to the distance from each neighbor candidate point to the current point, determining the maximum distance value in p points with the minimum distance; and taking all the neighbor candidate points of which the distances are smaller than or equal to the maximum distance value as neighbors of the current point. According to the method, flexible neighbor determination is realized, and a more accurate predicted value is provided during point cloud attribute prediction, so the coding and decoding performance of point cloud attributes is improved.

Description

technical field [0001] The present invention relates to the technical field of point cloud processing, in particular to a point cloud neighbor determination, point cloud prediction, point cloud encoding, and point cloud decoding method and device. Background technique [0002] 3D point cloud is an important form of digitalization of the real world. With the rapid development of 3D scanning equipment (such as laser, radar, etc.), the accuracy and resolution of point clouds have become higher. High-precision point clouds are widely used in the construction of urban digital maps, and play a technical supporting role in many popular researches such as smart cities, unmanned driving, and cultural relics protection. The point cloud is obtained by sampling the surface of an object by a 3D scanning device. The number of points in a frame of point cloud is generally in the millions, and each point contains geometric information, color, reflectivity and other attribute information, a...

AI Technical Summary

Problems solved by technology

[0004] In the existing technology, the neighbors at the same distance brought by the fixed number of neighbors are not used, and the distance brought by the fixed neighbor distance is too small to select neighbors or the distance is too large, so that some close neighbors are still selected. The problem of relatively weak long-distance neighbors cannot predict the attribute value of the current node well, which will easily lead to low attribute prediction accuracy, thereby reducing the encoding and decoding performance.

Images