DAPCC: Diverse Attention-Based Entropy Model for Dynamic LiDAR Point Cloud Compression

LiDAR point cloud (LPC) compression is an indispensable component for 3-D vision tasks, especially for dynamic point clouds. However, the existing methods based on traditional spatial-temporal attention (STA) are immature, causing little improvement in interframe feature extraction. In this article, we propose diverse attention-based point cloud compression (DAPCC), an LPC compression entropy model combining aggregation embedding (AggEmb) modules for temporal point matching and STA blocks for dynamic Octree node encoding, which can effectively utilize the change information of dynamic point clouds. Specifically, we first introduce AggEmb to match the Octree sequences from two sweeps to establish temporal correlation. To effectively capture the feature details, we further design local and global combined attention for the spatial-temporal information of point clouds which can focus on the whole context. Finally, we organize a symmetric multilayer perception (MLP) module capable of strengthening vital features. We conduct experiments of static and dynamic compression on both indoor/outdoor point cloud benchmark datasets (i.e., ScanNet, SemanticKITTI, and MPEG common test conditions (CTCs) Category 3 datasets) and downstream applications (i.e., vehicle detection and semantic segmentation). Compared with the previous state-of-the-art methods, our method achieves up to 14.7% bits per point (bpp) and 45% decoding time savings and adapts to the downstream tasks with almost no impact on performance.