Research map / Vision-language and multimodal / Cross-modal learning
p-AVAS: Can Physics-Integrated Audio-Visual Modeling Boost Neural Acoustic Synthesis?
Susan Liang, Chao Huang, Yunlong Tang, Zeliang Zhang, Chenliang Xu
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 vision-guided audio simulation as input | INRAS · RWAVS-Gen (overall) | MAG: 3.081 -> 2.763 | Improved |
| add vision-guided audio simulation as input | NAF · RWAVS-Gen (overall) | MAG: 2.900 -> 2.724 | Improved |
| add vision-guided audio simulation as input | ViGAS · RWAVS-Gen (overall) | MAG: 2.982 -> 2.438 | Improved |
| add vision-guided audio simulation as input | AV-NeRF · RWAVS-Gen (overall) | MAG: 3.168 -> 2.396 | Improved |
| add vision-guided audio simulation as input | AV-NeRF · RWAVS-Gen (overall) | ENV: 0.199 -> 0.163 | Improved |
| use mesh-based simulation instead of pyroomacoustics shoebox + HRTF (ablation) | pi-AVAS · RWAVS-Gen (overall), ablation | MAG: 5.435 -> 3.938 | Improved |
| use flow matching loss instead of regression loss for refinement (ablation) | pi-AVAS refinement network · RWAVS-Gen (overall), ablation | MAG: 2.844 -> 2.702 | Improved |
| add flow matching refinement to simulation (ablation) | pi-AVAS · RWAVS-Gen (overall), ablation | MAG: 3.938 -> 2.360 | Improved |
| add simulation stage to flow matching refinement (ablation) | pi-AVAS · RWAVS-Gen (overall), ablation | MAG: 2.702 -> 2.360 | Improved |
| add data augmentation (ablation) | pi-AVAS · RWAVS-Gen (overall), ablation | MAG: 2.360 -> 2.336 | Improved |
| add data augmentation on office and apartment scenes (ablation) | pi-AVAS · RWAVS-Gen office/apartment | MAG | No clear effect |
| add data augmentation on house and outdoor scenes (ablation) | pi-AVAS · RWAVS-Gen house/outdoor | MAG: 3.144 -> 1.946 (house) | Improved |
Closest papers by meaning
- AV-GS: Learning Material and Geometry Aware Priors for Novel View Acoustic Synthesis (NeurIPS 2024 · similarity 0.91)
- Acoustic Volume Rendering for Neural Impulse Response Fields (NeurIPS 2024 · similarity 0.89)
- Real Acoustic Fields: An Audio-Visual Room Acoustics Dataset and Benchmark (CVPR 2024 · similarity 0.89)
- NeRAF: 3D Scene Infused Neural Radiance and Acoustic Fields (ICLR 2025 · similarity 0.88)
- SoundVista: Novel-View Ambient Sound Synthesis via Visual-Acoustic Binding (CVPR 2025 · similarity 0.88)
- V2A-Mapper: A Lightweight Solution for Vision-to-Audio Generation by Connecting Foundation Models (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.
