Research map / Vision-language and multimodal / Cross-modal learning
Toward Interactive Sound Source Localization: Better Align Sight and Sound!
Arda Senocak, Hyeonggon Ryu, Junsik Kim, Tae-Hyun Oh, Hanspeter Pfister, Joon Son Chung
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 feature alignment (ablation) | Ours (NN w/ Sup. Pre. Enc.) · VGG-SS (ablation) | cIoU: 39.10 -> 39.94 | Improved |
| add multi-view invariance on top of semantic (ablation) | Ours (NN w/ Sup. Pre. Enc.) · VGG-SS (ablation) | cIoU: 38.75 -> 39.94 | Improved |
| add semantic invariance on top of multi-view (ablation) | Ours (NN w/ Sup. Pre. Enc.) · VGG-SS (ablation) | cIoU: 38.30 -> 39.94 | Improved |
| use semantic samples instead of multi-view samples (ablation) | Ours (NN w/ Sup. Pre. Enc.) · VGG-SS (ablation) | cIoU: 38.30 -> 38.75 | Improved |
| add all three components to baseline (ablation) | Ours (NN w/ Sup. Pre. Enc.) · VGG-SS (ablation) | cIoU: 34.22 -> 39.94 | Improved |
| add vision-modality hand-crafted positives (ablation) | Ours (NN w/ Sup. Pre. Enc.) · VGG-SS (ablation) | cIoU: 34.05 -> 37.72 | Improved |
| add audio-modality semantic positives (ablation) | Ours (NN w/ Sup. Pre. Enc.) · VGG-SS (ablation) | cIoU: 37.01 -> 38.75 | Improved |
| increase k in k-NN from 10 to 1000 (ablation) | Ours (NN w/ Sup. Pre. Enc.) · VGG-SS (ablation) | cIoU: 38.80 -> 39.94 | Improved |
| sample randomly from top-1000 instead of fixed same sample (k=1) (ablation) | Ours (NN w/ Sup. Pre. Enc.) · VGG-SS (ablation) | cIoU: 35.80 -> 38.24 | Improved |
| use pretrained vision encoder in NN search | Ours (NN w/ Sup. Pre. Enc.) · VGG-SS (w/o OGL) | cIoU: 39.94 -> 41.42 | Improved |
| use supervised instead of self-supervised pretrained encoder in NN search | Ours (NN w/ Sup. Pre. Enc.) · VGG-SS (w/o OGL) | cIoU: 39.16 -> 39.94 | Improved |
| add object-guided localization (OGL) to FNAC | FNAC · VGG-SS | cIoU: 39.50 -> 41.85 | Improved |
Closest papers by meaning
- Object-aware Sound Source Localization via Audio-Visual Scene Understanding (CVPR 2025 · similarity 0.91)
- T-VSL: Text-Guided Visual Sound Source Localization in Mixtures (CVPR 2024 · similarity 0.89)
- Seeing Sound, Hearing Sight: Uncovering Modality Bias and Conflict of AI models in Sound Localization (NeurIPS 2025 · similarity 0.89)
- Improving Sound Source Localization with Joint Slot Attention on Image and Audio (CVPR 2025 · similarity 0.88)
- Learning to Visually Localize Sound Sources from Mixtures without Prior Source Knowledge (CVPR 2024 · similarity 0.88)
- Robust Audio-Visual Segmentation via Audio-Guided Visual Convergent Alignment (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.
