Deep point cloud compression coding method based on full self-attention network

By constructing a point cloud compression method based on a full self-attention network and chamfer distance objective function training, the problems of insufficient utilization of point cloud sparsity and correlation are solved, and efficient point cloud compression and decompression are achieved, which is suitable for augmented reality, autonomous driving and other fields.

CN114363633BActive Publication Date: 2025-09-05SUN YAT SEN UNIV
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202111649443.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-12-30
Publication Date
2025-09-05
Estimated Expiration
2041-12-30

AI Technical Summary

Technical Problem

Among the existing learning-based point cloud compression methods, the voxel-based input method fails to fully consider the sparsity of point clouds, resulting in high complexity and difficulty in practical application; the point-based input method fails to fully utilize the correlation between points and has low encoding efficiency.

Method used

A point cloud compression encoding method based on a full self-attention network is constructed. By constructing a point cloud full self-attention network and using the chamfer distance objective function for training, the local and global correlations of the point cloud are learned, and the encoder and decoder are used for feature sampling and reconstruction to achieve point cloud compression and decompression.

Benefits of technology

It achieves point cloud compression with high coding efficiency and low complexity, is both effective and practical, can accurately represent point cloud semantic information, and ensure the security and stability of point cloud information storage and transmission.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114363633B_ABST
    Figure CN114363633B_ABST
Patent Text Reader

Abstract

The present invention discloses a method for deep point cloud compression encoding based on a full self-attention network. The method comprises: constructing a point cloud full self-attention network, the point cloud full self-attention network comprising an encoder and a decoder; acquiring training data, constructing a chamfer distance objective function to train the point cloud full self-attention network; inputting point cloud data into the trained point cloud full self-attention network, performing feature sampling processing on the point cloud data using the encoder to obtain point cloud encoding, and completing point cloud compression; and reconstructing the point cloud data using the decoder based on the point cloud encoding to complete point cloud decompression. The present invention strengthens the learning of local and global correlations between each point in the point cloud through network training based on the chamfer distance objective function, and samples the features of the point cloud through an encoder to obtain a point cloud encoding that can accurately represent the semantic information of the point cloud, while ensuring the security and stability of the storage and transportation of the point cloud information. The method can be widely applied in the field of point cloud compression encoding technology.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Point cloud geometric compression method based on depth auto-encoder

    CN110349230A

  • Point cloud geometric compression method based on deep convolutional network

    CN110691243A