Dual-stream cross-modality fusion transformer for RGB-D action recognition.
Zhen LiuJun ChengLibo LiuZiliang RenQieshi ZhangChengqun SongPublished in: Knowl. Based Syst. (2022)
Keyphrases
- action recognition
- depth cameras
- human actions
- human detection
- computer vision
- action classification
- activity recognition
- data streams
- bag of words
- video dataset
- recognizing human actions
- recognizing actions
- body parts
- static images
- multi sensor
- spatial temporal
- depth sensors
- bag of features
- depth information
- depth images
- recognition of human actions
- human activities
- action primitives
- human pose
- view invariant action recognition
- continuous queries
- depth map
- human computer interaction
- feature points
- text classification
- three dimensional