STA3D: Spatiotemporally attentive 3D network for video saliency prediction.
Wenbin ZouShengkai ZhuoYi TangShishun TianXia LiChen XuPublished in: Pattern Recognit. Lett. (2021)
Keyphrases
- visual attention
- multimedia
- computer networks
- network traffic
- peer to peer
- visual saliency
- video data
- prediction accuracy
- elman network
- video content
- video sequences
- network structure
- video frames
- saliency map
- radial basis function network
- network conditions
- video delivery
- video streams
- object recognition
- network model
- visual search
- salient regions
- video summarization
- neural network
- real time