Transtl: Spatial-Temporal Localization Transformer for Multi-Label Video Classification.
Hongjun WuMengzhu LiYongcheng LiuHongzhe LiuCheng XuXuewei LiPublished in: ICASSP (2022)
Keyphrases
- spatial temporal
- multi label
- video shots
- text categorization
- image classification
- image annotation
- graph cuts
- video retrieval
- visual features
- spatio temporal
- text classification
- temporal information
- semantic concepts
- video data
- class labels
- video content
- spatial and temporal
- key frames
- video analysis
- video database
- video sequences
- human actions
- video clips
- spatial information
- feature selection