Recur, Attend or Convolve? On Whether Temporal Modeling Matters for Cross-Domain Robustness in Action Recognition.
Sofia BrooméErnest PokropekBoyu LiHedvig KjellströmPublished in: WACV (2023)
Keyphrases
- action recognition
- cross domain
- spatial temporal
- bag of words
- activity recognition
- human actions
- action classification
- computer vision
- motion history images
- knowledge transfer
- transfer learning
- text categorization
- recognition of human actions
- spatio temporal
- temporal information
- recognizing actions
- human activities
- view invariant
- learning algorithm
- recognizing human actions
- active learning
- human object interactions
- action recognition in videos