Survey on Deep Learning-Based Point Cloud Compression

• DOI: 10.3389/frsip.2022.846972
• Publication Date: 2022-02-23T14:04:03Z
• Authors: Maurice Quach; Jiahao Pang; Dong Tian; Giuseppe Valenzise; Frederic Dufaux
• Keywords: quality assessment; 0202 electrical engineering, electronic engineering, information engineering; deep learning; rendering; Electrical engineering. Electronics. Nuclear engineering; 02 engineering and technology; compression; point cloud; TK1-9971

Point clouds are becoming essential in key applications with advances capture technologies leading to large volumes of data. Compression is thus for storage and transmission. In this work, the state art geometry attribute compression methods a focus on deep learning based approaches reviewed. The challenges faced when compressing attributes considered, an analysis current address them, their limitations relations between traditional ones. Current open questions point cloud compression, existing solutions perspectives identified discussed. Finally, link research problems relevant areas adjacent fields, such as rendering computer graphics, mesh quality assessment, highlighted.