Taeyun Woo

Ph.D. Student in Computer Science at KAIST (KAIST Computer Vision Lab)

DexHOI2Sim — Replicate MANO hand–object interactions in MuJoCo & IsaacGym and evaluate grasp success. A benchmark for HOI generation.

Hello! I am a Ph.D. student in Computer Science at KAIST, where I am fortunate to be co-advised by Tae-Kyun Kim and Jinah Park. Previously, I obtained my M.S. at KAIST advised by Jinah Park, after a B.S. in Mechanical Engineering (with a dual major in School of Computing) at KAIST. Since April 2026, I have also been a research scientist intern with the Embodied AI Team at NC AI, working on dexterous grasping of thin objects with the Unitree G1 humanoid and Dex3-1 hands.

My research enables AI to perceive and physically interact with the world through hands. I am particularly interested in grounding hand–object interaction in physics — bridging visual perception (hand pose estimation, hand–object reconstruction) with contact and force reasoning.