Papers
2
Total Citations
4
H-Index
1
About
Gaoqi He is a rising researcher at the forefront of human motion modeling and human-computer interaction, with a focus on generating realistic, socially aware, and emotionally expressive behaviors for virtual agents. His work bridges computer vision, graphics, and affective computing. He is best known for pioneering "Social-Scene-Aware Generative Adversarial Networks for Pedestrian Trajectory Prediction" (2021), a key contribution that models how individuals navigate crowded spaces by accounting for social context and scene semantics—a critical advance for autonomous driving and robotics. More recently, He has pushed the boundaries of embodied AI with "EmoDiffGes: Emotion-Aware Co-Speech Holistic Gesture Generation with Progressive Synergistic Diffusion" (2025). This work tackles the long-standing challenge of generating full-body, emotionally congruent gestures synchronized with speech, moving beyond simplistic or inexpressive motion. By leveraging diffusion models, his method produces nuanced, synergistic movements that reflect both the speaker's emotion and the speech rhythm. Though early in his career, He’s work is already garnering attention for its ambition to make virtual avatars and human-robot interaction more natural, empathetic, and socially intelligent.
Research Focus
Key Achievements
Top Papers
- 1
- 2