SOAR: Scene-debiasing Open-set Action Recognition.
Yuanhao ZhaiZiyi LiuZhenyu WuYi WuChunluan ZhouDavid S. DoermannJunsong YuanGang HuaPublished in: ICCV (2023)
Keyphrases
- action recognition
- motion features
- static images
- human actions
- activity recognition
- bag of words
- action classification
- d scene
- human detection
- computer vision
- recognition of human actions
- bag of features
- body parts
- video dataset
- scene classification
- depth sensors
- recognizing actions
- three dimensional
- single image
- video sequences
- recognizing human actions
- action detection
- view invariant
- visual words
- view invariant action recognition
- action primitives
- human activities
- scene structure
- moving objects
- atomic actions
- video images
- object recognition
- human pose
- image pairs
- image sequences
- action recognition in videos
- dynamic scenes