Huawen Liu

Papers

Total Citations

H-Index

About

Huawen Liu is a researcher whose work sits at the intersection of computer vision and human action analysis, with a particular focus on view-invariant recognition. His most-cited paper, "Dual-attention Network for View-invariant Action Recognition" (2023, 13 citations), addresses the persistent challenge of action occlusions and information loss caused by viewpoint changes—a critical problem for applications like visual surveillance and human-robot interaction. Liu's major contribution lies in developing attention-based architectures that can robustly recognize human actions regardless of camera perspective, effectively modeling spatiotemporal features that remain stable across different views. His work demonstrates a deep understanding of how to mitigate the distortions that plague traditional action recognition systems when the viewing angle shifts. While his citation count reflects a career still in its growth phase, the foundational nature of his research on dual-attention mechanisms positions him as an emerging voice in the field. For students and researchers exploring robust action recognition, Liu's work offers a clear pathway into solving one of computer vision's most practical and stubborn problems: making machines see and understand human movement as flexibly as humans do.