AdaFuse: Adaptive Temporal Fusion Network for Efficient Action Recognition.
Yue MengRameswar PandaChung-Ching LinPrasanna SattigeriLeonid KarlinskyKate SaenkoAude OlivaRogério FerisPublished in: ICLR (2021)
Keyphrases
- action recognition
- spatial temporal
- human actions
- activity recognition
- human detection
- computer vision
- static images
- recognizing human actions
- bag of words
- spatio temporal
- motion history images
- action classification
- human pose
- video dataset
- temporal information
- recognizing actions
- recognition of human actions
- space time
- independent subspace analysis
- bag of features
- body parts
- principal component analysis