We sense, model, and understand the geometric structure around us. The geometric structure stems from the physical space we live in, and we utilize the information to better combine knowledge and embed technology within the relationship between objects and humans. To do this, we adapt the state-of-the-art techniques from computer graphics, computer vision, machine learning, and robotics. The ultimate goal is to embrace the technology to seamlessly augment the limitation of time, space, or human, and make a better world.
Research Interests & Projects
1) 3D acquisition and modeling: With the help of various acquisition devices, we can capture and build a 3D model of objects, scenes, and human bodies. The research assists robots or human users to acquire 3D models of interest. We also ease the everyday use of acquired models in everyday life by updating the 3D models with light-weight acquisition devices or send the stream of 3D video over mobile networks.
2) Using 3D models for visualization: 3D models are crucial for seamless AR/VR applications and realistic rendering. The key technical components include localization, pose estimation, texture acquisition, and lighting estimation.
3) Using 3D data for perception: 3D information is widely used for perception, and recent trends on 3D perception utilize state-of-the-art techniques from computer vision and machine learning. We utilize neural networks of generative models, metric learning, and/or reinforcement learning to boost the performance.
4) Using 3D information for interaction: The real-world 3D models are utilized for manipulation, navigation, or other robotic applications. Our focus is to build algorithms to control a robot in un-constrained set-up and robust to small changes.