Procedural Generation of Videos to Train Deep Action Recognition Networks.
César Roberto de SouzaAdrien GaidonYohann CabonAntonio Manuel López PeñaPublished in: CoRR (2016)
Keyphrases
- action recognition
- human actions
- action classification
- recognition of human actions
- video dataset
- recognizing human actions
- ucf sports
- recognizing actions
- spatio temporal interest points
- human activities
- view invariant
- motion features
- static images
- bag of words
- action recognition in videos
- spatial temporal
- space time interest points
- activity recognition
- action detection
- human detection
- computer vision
- body parts
- bag of features
- motion capture data
- human object interactions
- action primitives
- video sequences
- mid level
- mid level features
- video frames
- motion history images
- low level descriptors
- visual features
- low level
- spatio temporal
- human pose