Multi-Level ResNets with Stacked SRUs for Action Recognition.
ZhenXing ZhengGaoyun AnQiuqi RuanPublished in: CoRR (2017)
Keyphrases
- action recognition
- bag of words
- human actions
- human detection
- activity recognition
- action classification
- computer vision
- spatial temporal
- recognizing actions
- body parts
- independent subspace analysis
- static images
- view invariant
- human pose
- mid level
- view invariant action recognition
- recognizing human actions
- action primitives
- space time interest points
- feature space
- motion features