Procedural Generation of Videos to Train Deep Action Recognition Networks.
César Roberto de SouzaAdrien GaidonYohann CabonAntonio Manuel López PeñaPublished in: CVPR (2017)
Keyphrases
- action recognition
- human actions
- recognition of human actions
- action classification
- video dataset
- recognizing human actions
- recognizing actions
- view invariant
- activity recognition
- static images
- ucf sports
- bag of words
- human activities
- motion features
- action detection
- spatio temporal interest points
- space time interest points
- human detection
- computer vision
- action recognition in videos
- spatial temporal
- human object interactions
- mid level features
- body parts
- video sequences
- bag of features
- mid level
- motion history images
- motion capture data
- low level descriptors
- multi view
- view invariant action recognition