Yu‐Hui Wen

Papers

Total Citations

171

H-Index

About

Yu-Hui Wen is a leading researcher in computer vision, robotics, and human motion synthesis, with a focus on enabling intelligent systems to perceive, plan, and interact naturally. Their work bridges active vision and generative motion modeling, addressing how robots can autonomously plan camera views to gather perceptual information efficiently—a critical capability for autonomous navigation and manipulation. Wen’s highly cited 2020 survey on view planning in robot active vision (109 citations) systematically reviews algorithms and systems, establishing a foundational reference that has shaped subsequent research in intelligent robotics. In human motion generation, Wen pioneered autoregressive stylized motion synthesis using generative flow models (39 citations), enabling unsupervised transfer of motion styles—a breakthrough for animation, gaming, and human-robot interaction. Their 2022 work on audio-driven stylized gesture generation (23 citations) further advances this line, creating realistic, synchronized gestures from speech inputs. Wen’s contributions have been recognized through publications in top venues like *Computer Graphics Forum* and *IEEE Transactions on Visualization and Computer Graphics*. By combining rigorous algorithmic development with practical applications, Wen continues to push the boundaries of how machines perceive and emulate human-like motion, making their work essential reading for students and researchers in embodied AI and computer graphics.