Research map / Video / Video-language understanding
Multi-modal Grouping Network for Weakly-Supervised Audio-Visual Video Parsing
Shentong Mo, Yapeng Tian
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 class-aware unimodal grouping (CUG) to baseline (ablation) | HAN baseline · LLP segment-level | F-score Visual: 52.9 -> 55.3 | Improved |
| Add CUG to baseline (ablation) | HAN baseline · LLP segment-level | F-score Audio: 60.1 -> 58.9 | Hurt |
| Add CUG to baseline (ablation) | HAN baseline · LLP segment-level | F-score Event@AV: 55.4 -> 55.9 | Improved |
| Add modality-aware cross-modal grouping (MCG) on top of CUG (ablation) | MGN · LLP segment-level | F-score Audio-Visual: 49.8 -> 50.6 | Improved |
| Add MCG on top of CUG (ablation) | MGN · LLP segment-level | F-score Type@AV: 54.7 -> 55.6 | Improved |
| Add MCG on top of CUG (ablation) | MGN · LLP segment-level | F-score Event@AV: 55.9 -> 57.2 | Improved |
| Use MGN instead of MA (w C) on audio-visual contrastive learning | MGN (w C) · LLP segment-level | F-score Visual: 53.1 -> 56.7 | Improved |
| Use MGN instead of MA with label refinement | MGN (w R) · LLP segment-level | F-score Visual: 57.5 -> 60.6 | Improved |
| Use MGN instead of MA with label refinement | MGN (w R) · LLP event-level | F-score Visual: 54.4 -> 58.4 | Improved |
| Use MGN (w C+R) instead of MA (w C+R) | MGN (w C+R) · LLP event-level | F-score Event@AV: 50.6 -> 49.9 | Hurt |
| Use MGN instead of HAN | MGN · LLP segment-level | F-score Audio: 60.1 -> 60.7 | No clear effect |
| Use MGN instead of HAN | MGN · LLP event-level | F-score Audio: 51.3 -> 51.0 | No clear effect |
Closest papers by meaning
- UWAV: Uncertainty-weighted Weakly-supervised Audio-Visual Video Parsing (CVPR 2025 · similarity 0.90)
- Weakly-Supervised Audio-Visual Video Parsing with Prototype-based Pseudo-Labeling (CVPR 2024 · similarity 0.90)
- Multimodal Class-aware Semantic Enhancement Network for Audio-Visual Video Parsing (AAAI 2025 · similarity 0.90)
- LongVALE: Vision-Audio-Language-Event Benchmark Towards Time-Aware Omni-Modal Perception of Long Videos (CVPR 2025 · similarity 0.88)
- Local-Global Multi-Modal Distillation for Weakly-Supervised Temporal Video Grounding (AAAI 2024 · similarity 0.87)
- Audio-Visual Semantic Graph Network for Audio-Visual Event Localization (CVPR 2025 · similarity 0.87)
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.
