A Prospective Study on Sequence-Driven Temporal Sampling and Ego-Motion Compensation for Action Recognition in the EPIC-Kitchens Dataset.
Alejandro López-CifuentesMarcos Escudero-ViñoloJesús BescósPublished in: CoRR (2020)
Keyphrases
- action recognition
- motion compensation
- human actions
- bag of words
- video compression
- spatial temporal
- video coding
- motion estimation
- motion vectors
- motion compensated
- body parts
- activity recognition
- video dataset
- multiscale
- action classification
- view invariant
- human detection
- computer vision
- recognizing human actions
- spatio temporal
- human motion
- block matching
- motion field
- image representation
- image quality