T-RECS: Training for Rate-Invariant Embeddings by Controlling Speed for Action Recognition.
Madan Ravi GaneshEric HofesmannByungsu MinNadha GafoorJason J. CorsoPublished in: CoRR (2018)
Keyphrases
- action recognition
- human actions
- activity recognition
- bag of words
- human detection
- computer vision
- body parts
- action classification
- recognizing actions
- spatial temporal
- recognizing human actions
- recognition of human actions
- depth sensors
- bag of features
- human pose
- action recognition in videos
- independent subspace analysis
- action detection
- view invariant action recognition
- view invariant
- video dataset
- human activities
- human body
- training set
- static images
- affine transformation
- image classification
- space time interest points
- feature extraction