Two-Streams: Dark and Light Networks with Graph Convolution for Action Recognition from Dark Videos (Student Abstract).
Saurabh SumanNilay NaharasBadri Narayan SubudhiVinit JakhetiyaPublished in: AAAI (2023)
Keyphrases
- action recognition
- human actions
- action classification
- video dataset
- recognition of human actions
- recognizing human actions
- space time interest points
- ucf sports
- spatio temporal interest points
- action recognition in videos
- human activities
- static images
- recognizing actions
- activity recognition
- action detection
- view invariant
- motion features
- bag of words
- human detection
- spatial temporal
- mid level features
- motion history images
- bag of features
- human object interactions
- depth sensors
- computer vision
- view invariant action recognition
- motion capture data
- high level
- graph structure
- body parts
- action primitives
- visual words
- video frames
- visual features
- sensor networks
- spatio temporal
- data streams
- video sequences