Generating Action-conditioned Prompts for Open-vocabulary Video Action Recognition.
Chengyou JiaMinnan LuoXiaojun ChangZhuohang DangMingfei HanMengmeng WangGuang DaiSizhe DangJingdong WangPublished in: CoRR (2023)
Keyphrases
- action recognition
- human actions
- action classification
- recognizing human actions
- action detection
- spatio temporal interest points
- space time interest points
- recognition of human actions
- atomic actions
- spatial temporal
- activity recognition
- bag of words
- view invariant
- recognizing actions
- static images
- action primitives
- computer vision
- human detection
- video dataset
- human activities
- human pose
- bag of features
- body parts
- space time
- motion history images
- mid level
- motion features
- spatio temporal
- action recognition in videos
- video content
- video shots
- human motion