Research map / Detection and segmentation / Weakly supervised detection and segmentation
Weakly-Supervised Audio-Visual Segmentation
Shentong Mo, Bhiksha Raj
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 |
|---|---|---|---|
| Use WS-AVS rather than C²AM | Weakly-supervised audio-visual segmentation · AVSBench single-source | mIoU: 30.87 -> 34.13 | Improved |
| Use WS-AVS rather than C²AM | Weakly-supervised audio-visual segmentation · AVSBench single-source | F-score: 36.55 -> 51.76 | Improved |
| Use WS-AVS rather than C²AM | Weakly-supervised audio-visual segmentation · AVSBench multiple-source | mIoU: 25.33 -> 30.85 | Improved |
| Use WS-AVS rather than C²AM | Weakly-supervised audio-visual segmentation · AVSBench multiple-source | F-score: 29.58 -> 46.87 | Improved |
| Add multi-scale audio-visual fusion with multiple-instance contrastive learning (ablation) | Weakly-supervised AVS baseline without AVF or PMR · AVSBench single-source | mIoU: 12.63 -> 31.85 | Improved |
| Add multi-scale audio-visual fusion with multiple-instance contrastive learning (ablation) | Weakly-supervised AVS baseline without AVF or PMR · AVSBench single-source | F-score: 24.99 -> 39.78 | Improved |
| Add pseudo-mask refinement using contrastive class-agnostic maps (ablation) | Weakly-supervised AVS baseline without AVF or PMR · AVSBench single-source | mIoU: 12.63 -> 32.46 | Improved |
| Add pseudo-mask refinement using contrastive class-agnostic maps (ablation) | Weakly-supervised AVS baseline without AVF or PMR · AVSBench single-source | F-score: 24.99 -> 43.92 | Improved |
| Add PMR alongside AVF (ablation) | AVF-only weakly-supervised AVS · AVSBench single-source | mIoU: 31.85 -> 34.13 | Improved |
| Add PMR alongside AVF (ablation) | AVF-only weakly-supervised AVS · AVSBench single-source | F-score: 39.78 -> 51.76 | Improved |
| Use four fusion stages rather than fewer stages (ablation) | WS-AVS · AVSBench single-source | mIoU | Improved |
| Increase M²ICL batch size from 8 to 64 (ablation) | WS-AVS · AVSBench single-source | mIoU | Improved |
Closest papers by meaning
- Unraveling Instance Associations: A Closer Look for Audio-Visual Segmentation (CVPR 2024 · similarity 0.92)
- Unsupervised Audio-Visual Segmentation with Modality Alignment (AAAI 2025 · similarity 0.89)
- Improving Audio-Visual Segmentation with Bidirectional Generation (AAAI 2024 · similarity 0.89)
- Audio-Visual Instance Segmentation (CVPR 2025 · similarity 0.88)
- Benchmarking Audio Visual Segmentation for Long-Untrimmed Videos (CVPR 2024 · similarity 0.88)
- Robust Audio-Visual Segmentation via Audio-Guided Visual Convergent Alignment (CVPR 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.
