Vision for Embodied Agent: Reconstruct and Understand the 3D world from the 2D images

​I’m currently Ph.D. candidate of National University of Singapore (NUS) under the supervision of Prof. Gim Hee Lee, conducting research in Robotics Perception. I'm strongly interested in building multi-modal AI, bridging the gap between natural language and various visual domains including images, 3D representations, and event streams. Specifically, my research interest is summarized as follows:
• Multimodality : VLM, 2D/3D-LLM
• (Open-World) Robotics : 3D Scene Understanding, Vision-and-Language Navigation, Object Manipulation
• Image Restoration : Motion Deblurring, Image Super Resolution

I'm serving as the chair of NUS Student Area Search Committees (ASC) for the Media research area, participating in the recruitment process of faculty members for the Department of Computer Science at NUS.

Before joining to NUS as Ph.D. candidate, I obtained the bachelor of Computer Science and Engineering in Korea University (KU). As an undergraduate research assistant, I had collaborated with various research institutions, mainly supervised by Prof. Gim Hee Lee (NUS), Prof. Angela Yao (NUS) and Prof. Seungryong Kim (KU, currently in KAIST). Thanks to their support, I was fortunate to have the opportunity to research open-world 2D/3D scene understanding and real-world image restoration as an undergraduate intern, leading to the publication of my paper at top-tier conferences such as ICLR and ECCV.