Research map / 3D vision / 3D generation
Visual-acoustic scene modeling
Synthesizes room acoustics and novel-view sound using simulation and learned acoustic fields.
12 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2022–2026) · an approach in 3D generation, keyword cluster “layout · indoor · 3d scene”
State History Tried Checklist Preprints Who Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
6 papers in 2023–24 and 5 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic shrank, from 1.3% to 0.6%.
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2022 · SoundSpaces 2.0: A Simulation Platform for Visual-Acoustic Learning (NeurIPS 2022 · 22 citations)
- 2023 · Novel-View Acoustic Synthesis (CVPR 2023 · 23 citations)
- 2024 · Hearing Anything Anywhere (CVPR 2024 · 8 citations)
- 2025 · Hearing Anywhere in Any Environment (CVPR 2025 · 6 citations)
- 2026 · Sonic4D: Spatial Audio Generation for Immersive 4D Scene Exploration (AAAI 2026 · 0 citations)
Earliest papers here
- SoundSpaces 2.0: A Simulation Platform for Visual-Acoustic Learning (NeurIPS 2022 · 22 citations)
- Novel-View Acoustic Synthesis (CVPR 2023 · 23 citations)
- AV-NeRF: Learning Neural Fields for Real-World Audio-Visual Scene Synthesis (NeurIPS 2023 · 6 citations)
What has been tried
11 papers here report 134 measured changes (a component added, removed or swapped, compared with a baseline): helped 69%, no clear effect 5%, hurt 24%, mixed 1%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| Add a learned source directivity pattern | DIFFRIR Dataset, four rooms · RIR Mag | Classroom: 5.22 -> 5.47; Dampened Room: 1.21 -> 1.64; Hallway: 9.13 -> 9.98; Complex Room: 4.86 -> 5.13 | Helped | Hearing Anything Anywhere (CVPR 2024 · 8 citations) |
| add visual input | Soundspace-Ambient seen scenes · STFT | 3.228 -> 2.442 | Helped | SoundVista: Novel-View Ambient Sound Synthesis via Visual-Acoustic Binding (CVPR 2025 · 3 citations) |
| replace pose-conditioned MLP baseline with AV-Cloud | RWAVS validation · MAG | 0.488 -> 0.351 | Helped | AV-Cloud: Spatial Audio Rendering Through Audio-Visual Cloud Splatting (NeurIPS 2024 · 3 citations) |
| increase the audio-token count from 1 to 5 | SoundBrush ablation evaluation · AVS | 0.235 -> 0.261 | Helped | SoundBrush: Sound as a Brush for Visual Scene Editing (AAAI 2025 · 1 citation) |
| Add ELE to AVR | GWA · STFT | 2.2 -> 1.7 | Helped | Resounding Acoustic Fields with Reciprocity (NeurIPS 2025 · 0 citations) |
| remove learnable room acoustic grid features | 17DRP5sb8fy and MeshRIR · overall RIR evaluation | Hurt | Deep Neural Room Acoustics Primitive (ICML 2024) |
What reviewers will expect
Datasets and baselines named in the 7 papers of this approach from 2024–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Baselines
- NAF · 4 of 7 papers
- INRAS · 3 of 7 papers
- AV-NeRF · 3 of 7 papers
- Nearest neighbor · 2 of 7 papers
- Linear interpolation · 2 of 7 papers
- Few-Shot RIR · 2 of 7 papers
- VAM · 2 of 7 papers
- ViGAS · 2 of 7 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 11 papers; from OpenAlex and the papers' affiliations).
- Meta · 5 papers
- University of Washington · 3 papers
- Stanford University · 2 papers
- University of Maryland, College Park · 2 papers
Use it
Download BibTeX of 12 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Generative indoor scene synthesis (131 papers)
- Open-vocabulary 3D scene understanding (95 papers)
Elsewhere in the topic
- Learning 3D reconstruction from images (196 papers)
- Diffusion-based novel view synthesis (156 papers)
- Semantic scene completion (153 papers)
- Hand-object interaction generation (116 papers)
- Feed-forward 3D asset generation and editing (116 papers)
- Score distillation for text-to-3D (106 papers)
- 2D-to-3D pretrained representations (94 papers)
- Monocular 3D object reconstruction (94 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).