From Isolated Islands to Pangea: Unifying Semantic Space for Human Action Understanding.
Yong-Lu LiXiaoqian WuXinpeng LiuYiming DouYikun JiJunyi ZhangYixing LiJingru TanXudong LuCewu LuPublished in: CoRR (2023)
Keyphrases
- human actions
- semantic space
- visual features
- action recognition
- spatio temporal
- human motion
- video sequences
- activity recognition
- co occurrence
- semantic concepts
- semantic features
- multi modal
- human activities
- cross modal
- image retrieval
- visual content
- automatically extracted
- latent semantic analysis
- visual information
- image classification
- computer vision
- space time
- feature space
- object recognition
- bag of features
- image sequences
- unsupervised manner
- high level