Huikun Bi

Institute of Computing Technology, Chinese Academy of Sciences

Papers

Total Citations

658

H-Index

About

Huikun Bi is a leading researcher in computer vision and autonomous systems, specializing in human trajectory prediction—a critical challenge for applications like autonomous driving and social robotics. Her work centers on modeling complex spatial-temporal interactions in crowded environments to enable safer, more intelligent navigation. Bi’s most influential contribution, the STGAT model (2019), has garnered over 630 citations for its innovative approach to capturing both spatial and temporal dynamics in pedestrian movement, significantly advancing prediction accuracy. She further extended this work with CoL-GAN (2020), an attention-based generative adversarial network that produces plausible, collision-free trajectories, addressing the dual challenges of realism and safety in dense crowds. Her research has been widely adopted in robotics and autonomous vehicle systems, where predicting human motion is essential for avoiding accidents. Bi’s achievements include pioneering deep learning frameworks that bridge the gap between theoretical modeling and real-world deployment, making her a key figure in the field. Her work continues to inspire new approaches to safe human-robot interaction and intelligent transportation.