Research map / Video / Video segmentation and tracking
Positive Sample Propagation Along the Audio-Visual Event Line
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use PSP to select and propagate high-similarity, including unsynchronized, audio-visual segment pairs (ablation) | AVE localization · AVE | accuracy: 73.7 -> 77.8; 70.2 -> 73.5 | Helped |
| filter weak and negative connections instead of propagating all connections (ablation) | PSP · AVE | accuracy: 75.9 -> 77.8; 71.2 -> 73.5 | Helped |
| filter weak connections as well as negative connections (ablation) | PSP · AVE | accuracy: 76.0 -> 77.8; 71.2 -> 73.5 | Helped |
| add self-attention before the Bi-LSTMs (ablation) | PSP · AVE | accuracy: 77.8 -> 75.4; 73.5 -> 70.8 | Hurt |
| add audio-visual pair similarity loss Lavps (ablation) | PSP · AVE fully-supervised | accuracy: 76.6 -> 77.8 | Helped |
| add a temporal weighting branch to the classification module (ablation) | PSP · AVE weakly-supervised | accuracy: 71.6 -> 73.5 | Helped |
| vary the similarity threshold τ (ablation) | PSP · AVE | accuracy: 75.9, 76.1, 75.3, 77.8, 76.6; 71.2, 71.7, 70.4, 73.5, 72.8 | Mixed |
Closest papers by meaning
- Towards Open-Vocabulary Audio-Visual Event Localization (CVPR 2025 · similarity 0.87)
- Unraveling Instance Associations: A Closer Look for Audio-Visual Segmentation (CVPR 2024 · similarity 0.87)
- Robust Audio-Visual Segmentation via Audio-Guided Visual Convergent Alignment (CVPR 2025 · similarity 0.86)
- Dense Audio-Visual Event Localization Under Cross-Modal Consistency and Multi-Temporal Granularity Collaboration (AAAI 2025 · similarity 0.86)
- Audio-Visual Semantic Graph Network for Audio-Visual Event Localization (CVPR 2025 · similarity 0.86)
- Weakly-Supervised Audio-Visual Video Parsing with Prototype-based Pseudo-Labeling (CVPR 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.