FitAQA: A Benchmark of Fitness Action Quality Assessment for Multimodal Large Language Models
FitAQA is introduced, a systematic benchmark for evaluating MLLMs in fitness AQA, containing 2,219 videos and 5,512 QA instances across 30 bodyweight exercises and a unified form error taxonomy is developed that defines 38 recurring form errors within six complementary quality dimensions.
Kai-Li Zheng, Kai-Wen Wang, Xun Zhu et al.
· 1 citation