Research map / Label-efficient and robust learning / Recognition and long-tail
A Closer Look at Weakly-Supervised Audio-Visual Source Localization
Shentong Mo, Pedro Morgado
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 visual dropout 0.9 (vs none) (ablation) | SLAVC · VGG-SS (extended) | AP: 26.03 -> 32.95 | Improved |
| increase visual dropout from 0.9 to 0.95 (ablation) | SLAVC · VGG-SS (extended) | AP: 32.95 -> 26.07 | Hurt |
| add audio dropout 0.5 (vs none) (ablation) | SLAVC · VGG-SS (extended) | AP: 32.95 -> 20.63 | Hurt |
| add momentum encoder 0.999 (vs none) (ablation) | SLAVC · VGG-SS (extended) | AP: 25.03 -> 32.95 | Improved |
| add SLAVC localization+correspondence factorization in training and inference (ablation) | SLAVC · VGG-SS (extended) | AP: 28.19 -> 32.95 | Improved |
| train with SLAVC but infer with AVLoc only (ablation) | SLAVC · VGG-SS (extended) | AP: 32.95 -> 22.01 | Hurt |
| add object-guided localization (OGL) to SLAVC (ablation) | SLAVC · VGG-SS (extended) | AP: 32.95 -> 34.46 | Improved |
| rule out early stopping for evaluation (EZ-VSL) | EZ-VSL · Flickr-SoundNet | LocAcc: 79.60 -> 66.40 | Hurt |
| rule out early stopping for evaluation (SLAVC) | SLAVC · Flickr-SoundNet | LocAcc: 83.20 -> 83.60 | No clear effect |
| use SLAVC instead of EZ-VSL without early stopping | EZ-VSL · Extended VGG-SS | AP: 24.55 -> 32.95 | Improved |
Closest papers by meaning
- Localizing Visual Sounds the Easy Way (ECCV 2022 · similarity 0.95)
- Localizing Visual Sounds the Hard Way (CVPR 2021 · similarity 0.94)
- Object-aware Sound Source Localization via Audio-Visual Scene Understanding (CVPR 2025 · similarity 0.92)
- T-VSL: Text-Guided Visual Sound Source Localization in Mixtures (CVPR 2024 · similarity 0.91)
- Prompting Segmentation with Sound Is Generalizable Audio-Visual Source Localizer (AAAI 2024 · similarity 0.91)
- Learning to Visually Localize Sound Sources from Mixtures without Prior Source Knowledge (CVPR 2024 · similarity 0.89)
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.
