Research map / Vision-language and multimodal / Cross-modal learning
Bootstrap Your Own AV-Proxies: Adaptive Contrastive and Prototype Learning for Audio-Visual Segmentation
Junbo Zhang, Hang Su, Zhaofan Li, Hang Dong, Chao Sun
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 SSAE module to baseline (ablation) | AVS baseline · AVS-Object-MS3 | J&F: 72.48 -> 76.22 | Improved |
| add DPC module to baseline (ablation) | AVS baseline · AVS-Object-MS3 | J&F: 72.48 -> 77.98 | Improved |
| add both SSAE and DPC (ablation) | AVS baseline · AVS-Semantic | J&F: 61.89 -> 69.61 | Improved |
| add SSAE on top of DPC (ablation) | DPC model · AVS-Object-MS3 | J&F: 77.98 -> 80.50 | Improved |
| add SSAE on top of DPC (ablation) | DPC model · AVS-Semantic | J&F: 67.40 -> 69.61 | Improved |
| use cross-attention only in SSAE (no GroupNorm) (ablation) | SSAE · AVS-Object-MS3 | J&F: 77.98 -> 77.54 | Hurt |
| use BatchNorm only in SSAE (ablation) | SSAE · AVS-Semantic | J&F: 67.40 -> 66.09 | Hurt |
| replace BatchNorm with GroupNorm after projection (with CA) (ablation) | SSAE · AVS-Object-MS3 | J&F: 78.85 -> 80.50 | Improved |
| replace BatchNorm with GroupNorm after projection (with CA) (ablation) | SSAE · AVS-Semantic | J&F: 68.76 -> 69.61 | Improved |
| add cross-attention before projection (with GN) (ablation) | SSAE · AVS-Object-MS3 | J&F: 77.77 -> 80.50 | Improved |
| GroupNorm only without cross-attention (ablation) | SSAE · AVS-Semantic | J&F: 67.40 -> 67.38 | No clear effect |
Closest papers by meaning
- Unraveling Instance Associations: A Closer Look for Audio-Visual Segmentation (CVPR 2024 · similarity 0.93)
- Dynamic Derivation and Elimination: Audio Visual Segmentation with Enhanced Audio Semantics (CVPR 2025 · similarity 0.90)
- Improving Audio-Visual Segmentation with Bidirectional Generation (AAAI 2024 · similarity 0.89)
- Unsupervised Audio-Visual Segmentation with Modality Alignment (AAAI 2025 · similarity 0.89)
- AVSegFormer: Audio-Visual Segmentation with Transformer (AAAI 2024 · similarity 0.89)
- Audio-Visual Instance Segmentation (CVPR 2025 · 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.
