Papers
39
Total Citations
700
H-Index
9
About
Minsu Jang is a versatile robotics and human-computer interaction researcher whose work spans gesture generation, activity recognition, and social robotics. Perhaps his most influential contribution is his 2020 paper on trimodal co-speech gesture generation — drawing on text, audio, and speaker identity — which has amassed over 300 citations and represents a landmark advance in making virtual avatars and social robots behave more naturally during conversation. Complementing this, his earlier end-to-end learning approach for humanoid robot gesture generation (2019) helped lay the groundwork for machine-learned, rather than rule-based, social behaviors in robots. Jang has also made significant strides in dataset development, introducing ETRI-Activity3D, a large-scale RGB-D dataset for recognizing elderly daily activities (82 citations), and AIR-Act2Act, a human-human interaction dataset for teaching robots non-verbal social behaviors. His review papers on personalization in human-robot interaction and robots in museum settings reflect a broader commitment to translating technical methods into real-world applications. Reaching back further, his foundational work on ubiquitous robotic spaces and space robotics mechatronics demonstrates a career-long dedication to pushing the boundaries of intelligent, context-aware robotic systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6Design and Implementation of a Ubiquitous Robotic Space25 citations · 2009
- 7
- 8Building an automated engagement recognizer based on video analysis13 citations · 2014
- 9
- 10Ubiquitous robot simulation framework and its applications9 citations · 2005