Contrastive Loss Based Frame-Wise Feature Disentanglement for Polyphonic Sound Event Detection.
Yadong GuanJiqing HanHongwei SongWenjie SongGuibin ZhengTieran ZhengYongjun HePublished in: ICASSP (2024)
Keyphrases
- event detection
- video analysis
- activity recognition
- video surveillance
- event recognition
- video event detection
- surveillance videos
- pairwise
- spatio temporal
- image features
- scan statistic
- musical instrument
- sports video
- video event
- text streams
- video frames
- complex events
- musical instruments
- video sequences
- semantic event detection