Research map / Detection and segmentation / Open-set and few-shot detection
A Proposal-Based Paradigm for Self-Supervised Sound Source Localization in Videos
Hanyu Xuan, Zhiliang Wu, Jian Yang, Yan Yan, Xavier Alameda-Pineda
Check 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 |
|---|---|---|---|
| Use proposal-based localization instead of the strongest listed map-based baseline | Sound source localization · MUSIC-solo | IoU@0.5: 52.3 -> 58.0 | Improved |
| Use proposal-based localization instead of the strongest listed map-based baseline | Sound source localization · MUSIC-Synthetic | IoU@0.5: 40.6 -> 46.7 | Improved |
| Use proposal-based localization instead of the strongest listed map-based baseline | Sound source localization · MUSIC-Synthetic | AUC: 23.6 -> 30.1 | Improved |
| Use proposal-based localization instead of the strongest listed map-based baseline | Sound source localization · SSL | cIoU: 69.9 -> 72.8 | Improved |
| Use proposal-based localization instead of the strongest listed map-based baseline | Sound source localization · SSL | AUC: 57.3 -> 61.4 | Improved |
| Use K=10 proposals instead of K=20 (ablation) | Proposal-based localization · SSL | cIoU: 72.8 -> 68.9 | Hurt |
| Use K=10 proposals instead of K=20 (ablation) | Proposal-based localization · SSL | AUC: 61.4 -> 56.2 | Hurt |
| Use K=30 proposals instead of K=20 (ablation) | Proposal-based localization · SSL | cIoU: 72.8 -> 73.1 | Improved |
| Use K=30 proposals instead of K=20 (ablation) | Proposal-based localization · SSL | AUC: 61.4 -> 57.3 | Hurt |
| Add GRM-guided Proposal Selection Network rather than search all proposals (ablation) | Proposal-based localization · SSL | cIoU: 58.9 -> 72.8 | Improved |
| Add GRM-guided Proposal Selection Network rather than search all proposals (ablation) | Proposal-based localization · SSL | AUC: 48.5 -> 61.4 | Improved |
| Use proposal-based localization instead of the location map directly (ablation) | Sound source localization · SSL | cIoU: 69.2 -> 72.8 | Improved |
Closest papers by meaning
- Robust Audio-Visual Contrastive Learning for Proposal-Based Self-Supervised Sound Source Localization in Videos (TPAMI 2024 · similarity 0.92)
- Whence the Voice? Self-supervised Dual-source Audio-Visual Localisation via Selective Convergence (ECCV 2026 · similarity 0.90)
- Learning to Localize Sound Sources in Visual Scenes: Analysis and Applications (TPAMI 2019 · similarity 0.90)
- Learning to Visually Localize Sound Sources from Mixtures without Prior Source Knowledge (CVPR 2024 · similarity 0.89)
- Object-aware Sound Source Localization via Audio-Visual Scene Understanding (CVPR 2025 · similarity 0.88)
- Prompting Segmentation with Sound Is Generalizable Audio-Visual Source Localizer (AAAI 2024 · 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.
