KSF-ST: Video Captioning Based on Key Semantic Frames Extraction and Spatio-Temporal Attention Mechanism.
Zhaowei QuLuhan ZhangXiaoru WangBingyu CaoYueli LiFu LiPublished in: IWCMC (2020)
Keyphrases
- spatio temporal
- video frames
- attention mechanism
- spatial and temporal
- temporal domain
- key frames
- spatio temporally
- moving objects
- saliency map
- space time
- spatial and temporal relationships
- motion trajectories
- video sequences
- visual attention
- sports video
- video data
- video content
- human actions
- temporal filtering
- image sequences
- video objects
- video streams
- video clips
- real time
- video surveillance
- low level features
- background subtraction
- video analysis
- eye movements
- computational complexity
- high level
- image processing
- semantic information