CSTA: CNN-based Spatiotemporal Attention for Video Summarization.
Jaewon SonJaehun ParkKwangsu KimPublished in: CoRR (2024)
Keyphrases
- video summarization
- audio visual
- video content
- video browsing
- video data
- video summaries
- event detection
- video sequences
- spatio temporal
- surveillance videos
- moving objects
- key frames
- video retrieval
- video frames
- space time
- visual attention
- personal photos
- real time
- low level features
- similarity measure
- three dimensional
- computer vision
- information retrieval