Reproduce “Attribute-Centric Vision-Language Alignment for Fine-Grained Gait Understanding” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Saihui Hou, Chenye Wang, Wenpeng Lang, Qiannan Guo, Zhengxiang Lan, Chao Fan, Shiqi Yu, Yongzhen Huang
The authors introduce Gait-VL, which matches walking sequences to text descriptions of gait attributes. It uses attribute-specific queries to gather visual evidence across frames and image locations, then compares it with text prototypes. On RA-GAR, Gait-VL reaches 73.28% mA, versus 71.19% for CLIP-GAR; this establishes performance on the benchmark, not universal reliability.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.