Three-Stream Network With Bidirectional Self-Attention for Action Recognition in Extreme Low Resolution Videos.
Didik PurwantoRizard Renanda Adhi PramonoYie-Tarng ChenWen-Hsien FangPublished in: IEEE Signal Process. Lett. (2019)
Keyphrases
- action recognition
- low resolution
- human actions
- high resolution
- action classification
- recognition of human actions
- super resolution
- recognizing human actions
- view invariant
- low resolution images
- human activities
- action recognition in videos
- recognizing actions
- space time interest points
- bag of words
- activity recognition
- low frame rate
- computer vision
- video sequences
- image patches
- high quality
- face images
- human pose
- video frames
- data streams
- machine learning
- higher resolution
- depth images
- color images
- depth map
- spatial resolution