Your Camera Improves Your Point Cloud Compression

LiDAR point cloud compression is important for autonomous driving as it consumes a lot of storage and bandwidth. Al- though the fusion of camera and LiDAR for vision percep- tion has been well studied, it remains unexplored that how we can improve the compression of LiDAR point cloud data using cross-modal information from cameras. In this paper, we propose a multi-modality compression framework for Li- DAR point cloud by exploiting the depth information pre- dicted from its paired image. To the best of our knowledge, our model is the first multi-modality compression framework for point cloud. Specifically, we first represent point cloud based on octrees to reduce spatial redundancy. Then, we pro- pose a cross-modal fusion structure to improve the compres- sion of these octrees, with depth distribution extracted from the camera pixels and acts as side information. Compared to previous state-of-the-art (SOTA) method, our approach ob- tains up to 8.10% compression rate gain for LiDAR point cloud compression.