Papers
17
Total Citations
400
H-Index
6
About
Joo-Haeng Lee is a leading researcher at the intersection of human-robot interaction, spatial augmented reality, and embodied AI. His work centers on making robots and virtual agents more intuitive and expressive, with major contributions in co-speech gesture generation, motion retargeting, and modular manipulation systems. His most influential paper, "Speech gesture generation from the trimodal context of text, audio, and speaker identity" (2020, 300+ citations), pioneered a method for generating human-like gestures for avatars and social robots, significantly advancing the naturalness of human-agent interaction. Lee also developed C-3PO, a cyclic three-phase optimization framework using reinforcement learning for human-robot motion retargeting, enabling kinematic-independent control across heterogeneous robot morphologies. His research on robotic spatial augmented reality (RSAR) systems, which combine robotics with projectors and cameras to augment real objects, has been foundational for interactive workspaces and cooperative activities. Lee’s notable achievements include the development of the MODMAN self-reconfigurable modular manipulation system and the Future Robotic Computer (FRC) concept, which integrates cameras, projectors, and actuators for dynamic human-robot collaboration. With a strong focus on data-driven kinematic control and curriculum learning, his work continues to shape the future of socially aware and physically capable robotic systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6A note on hybrid control of robotic spatial augmented reality7 citations · 2011
- 7
- 8FRC based augment reality for aiding cooperative activities5 citations · 2013
- 9
- 10Issues in Control of a Robotic Spatial Augmented Reality System5 citations · 2011