Papers
8
Total Citations
93
H-Index
6
About
Kalin Stefanov is a leading researcher in multimodal human-machine interaction, specializing in the analysis and modeling of nonverbal communication signals such as hand gestures, speech activity, and eye-gaze. His work bridges social robotics and computational linguistics, with a focus on enabling machines to understand and generate natural social behaviors. Stefanov’s major contributions include developing self-supervised methods for active speaker detection in multiparty settings—a fundamental step for socially aware language acquisition—and pioneering the analysis of intimate self-disclosure through verbal and nonverbal cues, demonstrating that computers can support mental well-being. His highly cited 2019 paper on multimodal analysis of self-disclosure (29 citations) and his 2019 work on self-supervised active speaker detection (18 citations) underscore his impact. Stefanov also created the “Look who’s talking” multimodal dataset for studying visual attention in human-human and human-robot interaction, and contributed to the development of the Furhat robot head for multiparty conversational systems. His research, with over 90 total citations, lays essential groundwork for building empathetic, socially aware artificial agents capable of natural, face-to-face communication.
Research Focus
Key Achievements
Top Papers
- 1Multimodal Analysis and Estimation of Intimate Self-Disclosure29 citations · 2019
- 2
- 3Look who's talking12 citations · 2016
- 4
- 5Tutoring Robots8 citations · 2014
- 6Multimodal multiparty social interaction with the furhat head8 citations · 2012
- 7
- 8