Research map / Video / Video segmentation and tracking
Exploring Cross-Video and Cross-Modality Signals for Weakly-Supervised Audio-Visual Video Parsing
Yan-Bo Lin, Hung-Yu Tseng, Hsin-Ying Lee, Yen-Yu Lin, Ming-Hsuan Yang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| add cross-video shared semantics (CM-S) (ablation) | HAN · LLP test | visual segment F-score: 52.9 -> 58.3 | Improved |
| add cross-video shared semantics (CM-S) (ablation) | HAN · LLP test | audio segment F-score: 60.1 -> 58.1 | Hurt |
| add cross-modality co-occurrence (CM-Co) (ablation) | HAN · LLP test | Event@AV segment F-score: 55.4 -> 57.4 | Improved |
| add CM-S (ablation) | MA* · LLP test | visual segment F-score: 57.5 -> 60.7 | Improved |
| add CM-Co (ablation) | MA* · LLP test | Event@AV segment F-score: 56.6 -> 59.1 | Improved |
| add both CM-S and CM-Co (ablation) | MA* · LLP test | visual segment F-score: 57.5 -> 63.5 | Improved |
| add both CM-S and CM-Co (ablation) | HAN · LLP test | Type@AV segment F-score: 54.0 -> 57.5 | Improved |
| co-occurrence on visual classes only instead of cross-modality co-attention (ablation) | MA* · LLP test | audio segment F-score: 60.5 -> 46.8 | Hurt |
| add cross-modality co-attention on top of within-modality self-attention (ablation) | MA* · LLP test | Event@AV segment F-score: 56.7 -> 59.1 | Improved |
| co-occurrence on audio classes only (ablation) | MA* · LLP test | audio segment F-score: 59.8 -> 60.7 | Improved |
| add more transformer layers to HAN instead of CM-Co (ablation) | HAN · LLP test | Event@AV segment F-score: 55.4 -> 54.9 | No clear effect |
| add CM-Co module vs adding a layer (ablation) | HAN · LLP test | Event@AV segment F-score: 54.9 -> 57.4 | Improved |
Closest papers by meaning
- Weakly-Supervised Audio-Visual Video Parsing with Prototype-based Pseudo-Labeling (CVPR 2024 · similarity 0.93)
- Multimodal Class-aware Semantic Enhancement Network for Audio-Visual Video Parsing (AAAI 2025 · similarity 0.91)
- UWAV: Uncertainty-weighted Weakly-supervised Audio-Visual Video Parsing (CVPR 2025 · similarity 0.91)
- PreFM: Online Audio-Visual Event Parsing via Predictive Future Modeling (NeurIPS 2025 · similarity 0.90)
- Audio-Visual Semantic Graph Network for Audio-Visual Event Localization (CVPR 2025 · similarity 0.88)
- Dense Audio-Visual Event Localization Under Cross-Modal Consistency and Multi-Temporal Granularity Collaboration (AAAI 2025 · similarity 0.88)
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.
