DC3D: A Video Action Recognition Network Based on Dense Connection.
Xiaofang MuZhenyu LiuJiaji LiuHao LiYue LiYikun LiPublished in: CBD (2022)
Keyphrases
- action recognition
- human actions
- action classification
- video dataset
- spatial temporal
- action detection
- recognition of human actions
- bag of words
- motion features
- recognizing human actions
- static images
- space time interest points
- human activities
- activity recognition
- computer vision
- human detection
- video data
- video sequences
- body parts
- recognizing actions
- mid level
- motion history images
- video frames
- view invariant
- multimedia
- action recognition in videos
- action primitives
- motion capture data
- video content
- visual features
- depth sensors
- video streams
- human pose
- video clips
- object detection
- event recognition
- key frames
- image sequences