Research map / Vision-language and multimodal / Cross-modal learning
Vision-Infused Deep Audio Inpainting
Hang Zhou, Ziwei Liu, Xudong Xu, Ping Luo, Xiaogang Wang
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 audio-visual synchronizing loss (ablation) | VIAI-AV · MUSICES audio inpainting | PSNR: 21.8 -> 23.2 | Improved |
| add audio-visual synchronizing loss (ablation) | VIAI-AV · MUSICES audio inpainting | SSIM: 0.62 -> 0.64 | Improved |
| add probe loss (ablation) | VIAI-AV · MUSICES audio inpainting | PSNR: 22.5 -> 23.2 | Improved |
| add probe loss (ablation) | VIAI-AV · MUSICES audio inpainting | SSIM: 0.63 -> 0.64 | Improved |
| use adaptive weight eta1(t) instead of fixed eta1(0) (ablation) | VIAI-A · MUSICES audio inpainting | PSNR: 21.8 -> 22.2 | Improved |
| use adaptive weight eta1(t) instead of fixed eta1(inf) (ablation) | VIAI-A · MUSICES audio inpainting | PSNR: 21.6 -> 22.2 | Improved |
| use interpolation initialization instead of fixed-value initialization (ablation) | VIAI-A · MUSICES audio inpainting | PSNR: 21.5 -> 22.2 | Improved |
| add WaveNet conditioning on past results (ablation) | VIAI-AV · MUSICES user study | audio quality MOS: 3.00 -> 3.93 | Improved |
| add synchronizing loss (user study) (ablation) | VIAI-AV · MUSICES user study | audio quality MOS: 2.90 -> 3.93 | Improved |
| add visual information to the inpainter | VIAI-A to VIAI-AV · MUSICES audio inpainting | PSNR: 22.2 -> 23.2 | Improved |
| add visual information to the inpainter | VIAI-A to VIAI-AV · MUSICES audio inpainting | OPS: 52.4 -> 56.3 | Improved |
| add visual information to the inpainter | VIAI-A to VIAI-AV · MUSICES audio inpainting | SDR: 6.54 -> 6.63 | No clear effect |
Closest papers by meaning
- Compose with Me: Collaborative Music Inpainter for Symbolic Music Infilling (AAAI 2025 · similarity 0.86)
- X-AVDD: Cross-Attentive Audio-Visual Dataset Distillation (NeurIPS 2026 · similarity 0.85)
- SoundBrush: Sound as a Brush for Visual Scene Editing (AAAI 2025 · similarity 0.85)
- From Vision to Audio and Beyond: A Unified Model for Audio-Visual Representation and Generation (ICML 2024 · similarity 0.85)
- Learning to Highlight Audio by Watching Movies (CVPR 2025 · similarity 0.85)
- Dynamic Derivation and Elimination: Audio Visual Segmentation with Enhanced Audio Semantics (CVPR 2025 · similarity 0.84)
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.
