Temporal Attention Mechanism with Conditional Inference for Large-Scale Multi-label Video Classification.
Eun-Sol KimKyoung-Woon OnJongseok KimYu-Jung HeoSeong-Ho ChoiHyun-Dong LeeByoung-Tak ZhangPublished in: ECCV Workshops (4) (2018)
Keyphrases
- multi label
- video classification
- attention mechanism
- multi label classification
- text categorization
- image classification
- image annotation
- graph cuts
- visual attention
- text classification
- dynamic textures
- temporal information
- saliency map
- video clips
- video content
- spatio temporal
- video shots
- classifier training
- spatial and temporal
- class labels
- space time
- motion trajectories
- knn
- image processing
- markov random field
- multiscale
- high level
- image segmentation