Yu–Gang Jiang

Fudan University, Shanghai Key Laboratory of Trustworthy Computing

Papers

Total Citations

H-Index

About

Yu-Gang Jiang is a leading figure in computer vision and multimedia computing, renowned for his pioneering work in large-scale video understanding, generative AI, and robotic perception. His research spans from foundational problems in video content recognition to cutting-edge applications in instruction-guided video prediction and robotic manipulation. A key contribution is his leadership in the LSVC2017 challenge, which advanced the field of large-scale video classification by addressing the critical challenge of recognizing visual content in unconstrained videos—a task vital for web search, advertising, and robotics. With over 3,000 citations, his work has had a profound impact. Notably, his recent innovations include "Aid," a method adapting diffusion models for text-guided video prediction, enabling applications in virtual reality and content creation, and a unified, real-time approach for category-level 6D pose estimation of articulated objects, directly improving robotic grasping. These achievements underscore his ability to bridge fundamental research with practical, real-world systems, making him a highly influential voice in the AI community.