Our goal is to capture, perceive and understand the human, scenes, and the interactions between humans and scenes in the complex real world. Our research topics include scene understanding and 3D reconstruction, autonomous driving, multi-modal learning, point cloud perception and representation, human motion capture/generation/prediction. etc. Our research has vital applications in assistive robotics, improving autonomous driving system safety and functionality, advancing surveillance in various environment, and enhancing human-robot cooperation. We aim to make a substantial impact on technological advancement, enriching our interactions with and comprehension of the complex world.