🤖 Byeongjin Kang
Ph.D. Student @ Yonsei University | 3D Vision & Embodied AI
I am a Ph.D. student in Artificial Intelligence at Yonsei University,
in the V-Lab advised by Prof. EunByung Park.
My research focuses on 3D vision and robot learning, developing AI systems that perceive and understand the real world through 3-dimensional spatial reasoning from multimodal sensor data. I aim to create frameworks where 3D perception enables intelligent robotic behavior to solve complex real-world problems.
Previously, Research intern at RLLAB (Prof. Youngwoon Lee) and CSI Lab (Prof. Yusung Kim).
[ Publications ]
Coming soon...
[ Projects & Activities ]
This project explored online reinforcement learning for a complex horizontal task using a main board and cables. While full success wasn't achieved, it provided valuable insights into real-time adaptation challenges and established groundwork for future online RL advancements.
This project enhances robotic manipulation through imitation learning with visual robustness. Using multiple cameras and vision techniques, it handles lighting and color variations effectively. The ACT model achieved 10% success in training environments and 80% across diverse test conditions.
This project develops a novel segmentation technique using cosine similarity and clustering for video lecture summarization. A compact 500M-parameter LLM then summarizes these segments, outperforming baselines with reduced computational cost.