Combining Per-frame and Per-track Cues for Multi-person Action Recognition.
Sameh KhamisVlad I. MorariuLarry S. DavisPublished in: ECCV (1) (2012)
Keyphrases
- action recognition
- mid level
- bag of words
- human actions
- activity recognition
- computer vision
- recognizing human actions
- spatial temporal
- body parts
- action classification
- human detection
- static images
- independent subspace analysis
- human pose
- recognition of human actions
- human activities
- action primitives
- motion features
- video dataset
- action recognition in videos
- recognizing actions
- view invariant
- action detection
- machine learning
- frame rate
- visual features