SRI3D: Two-stream inflated 3D ConvNet based on sparse regularization for action recognition.
Zhaoqilin YangGaoyun AnRuichen ZhangZhenxing ZhengQiuqi RuanPublished in: IET Image Process. (2023)
Keyphrases
- action recognition
- activity recognition
- human actions
- bag of words
- human detection
- action classification
- data streams
- body parts
- computer vision
- spatial temporal
- static images
- human activities
- bag of features
- recognition of human actions
- human activity recognition
- view invariant
- view invariant action recognition
- action recognition in videos
- recognizing actions
- depth sensors
- action detection
- stream processing
- sparse representation
- machine learning
- max margin
- multi class