I expect the depth map to use camera information and AI inference as well as LiDAR.