Spatiotemporal module for video saliency prediction based on self-attention.
Yuhao WangZhuoran LiuYibo XiaChunbo ZhuDanpei ZhaoPublished in: Image Vis. Comput. (2021)
Keyphrases
- visual saliency
- visual attention
- saliency map
- space time
- spatial and temporal
- video frames
- prediction accuracy
- video sequences
- spatiotemporal features
- spatio temporal
- video data
- video content
- saliency detection
- multimedia
- attention mechanism
- natural images
- video database
- spatiotemporal patterns
- dynamic textures
- prediction model
- video streams
- bit rate
- key frames
- prediction algorithm
- salient regions
- scalable video coding
- video summarization
- eye movements
- eye tracking
- salient points
- multimedia data
- video surveillance
- focus of attention
- digital video
- video segmentation
- input image
- prediction error
- motion estimation