ViGAT: Bottom-up event recognition and explanation in video using factorized graph attention network.
Nikolaos GkalelisDimitrios DaskalakisVasileios MezarisPublished in: CoRR (2022)
Keyphrases
- event recognition
- event detection
- video analysis
- video event detection
- multi camera
- visual attention
- video event
- complex events
- activity recognition
- video surveillance
- web videos
- trecvid multimedia event detection
- event models
- video data
- motion estimation
- video processing
- concept detectors
- video clips
- video annotation
- image processing
- surveillance system
- video streams
- object tracking
- mean shift
- high resolution