Papers
4
Total Citations
51
H-Index
4
About
He Cao is a researcher at the forefront of embodied AI and robotic perception, with a core focus on deep reinforcement learning, multi-modal fusion, and human-robot interaction. His work bridges the gap between high-level visual understanding and low-level robotic control, particularly in cluttered, dynamic environments. Cao’s most impactful contribution is the development of an efficient hybrid-supervised deep reinforcement learning framework for person-following robots (21 citations), which significantly improved the robustness of autonomous tracking. He has also advanced affective computing through a weakly supervised facial expression recognition system that leverages transferred deep active learning (18 citations). More recently, Cao has pioneered multi-modal fusion architectures for robotic manipulation, introducing a joint segmentation and grasp pose detection network that integrates point clouds and RGB images to enhance performance in cluttered scenes. His transformer-based RGB-D fusion network for desktop object instance segmentation further demonstrates his commitment to leveraging state-of-the-art attention mechanisms for precise scene understanding. With a growing citation footprint, He Cao is establishing himself as a key innovator in making robots more perceptive, adaptive, and capable of operating safely alongside humans.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4