Research map / 3D vision / 3D generation
Differentiable Room Acoustic Rendering with Multi-View Vision Priors
Derong Jin, Ruohan Gao
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 |
|---|---|---|---|
| replace beam tracing with specular ray tracing (ablation) | AV-DAR · RAF-Furnished (0.1% data) | C50 error (dB): 1.98 -> 4.27 | Hurt |
| add position-dependent residual component (ablation) | AV-DAR · RAF-Furnished (0.1% data) | T60 error (%): 49.0 -> 15.2 | Improved |
| make residual position-dependent instead of a uniform learnable vector (ablation) | AV-DAR · RAF-Furnished (0.1% data) | EDT error (ms): 106.4 -> 80.1 | Improved |
| make residual position-dependent instead of uniform (ablation) | AV-DAR · RAF-Furnished (0.1% data) | T60 error (%): 13.9 -> 15.2 | Hurt |
| add multi-view vision features (ablation) | AV-DAR · RAF-Furnished (0.1% data) | EDT error (ms): 98.6 -> 80.1 | Improved |
| add vision features (ablation) | AV-DAR · RAF-Furnished (0.1% data) | T60 error (%): 14.3 -> 15.2 | No clear effect |
| use integrated positional encoding (IPE) (ablation) | AV-DAR · RAF-Furnished (0.1% data) | EDT error (ms): 101.2 -> 80.1 | Improved |
| use DINOv2 instead of ResNet18 encoder (ablation) | AV-DAR · RAF-Furnished (0.1% data) | EDT error (ms): 89.4 -> 80.1 | Improved |
| use 20 views instead of 10 views for vision features (ablation) | AV-DAR · RAF-Furnished (0.1% data) | EDT error (ms): 97.9 -> 80.9 | Improved |
| use 65 views instead of 20 views (ablation) | AV-DAR · RAF-Furnished (0.1% data) | EDT error (ms): 80.9 -> 80.1 | No clear effect |
| AV-DAR vs baselines at equal 1% training data | AV-DAR · RAF | relative error reduction: 16.6% to 50.9% | Improved |
| AV-DAR vs AV-NeRF on narrow hallway scene | AV-DAR · HAA Hallway | C50 error (dB): 1.03 -> 1.15 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.
- AV-NeRF — AV-NeRF: Learning Neural Fields for Real-World Audio-Visual Scene Synthesis (NeurIPS 2023) · better on 1 of 2 shared results
- INRAS++ — INRAS: Implicit Neural Representation for Audio Scenes (NeurIPS 2022) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Acoustic Volume Rendering for Neural Impulse Response Fields (NeurIPS 2024 · similarity 0.88)
- Hearing Anything Anywhere (CVPR 2024 · similarity 0.88)
- Real Acoustic Fields: An Audio-Visual Room Acoustics Dataset and Benchmark (CVPR 2024 · similarity 0.88)
- AV-GS: Learning Material and Geometry Aware Priors for Novel View Acoustic Synthesis (NeurIPS 2024 · similarity 0.88)
- Hearing Anywhere in Any Environment (CVPR 2025 · similarity 0.86)
- Resounding Acoustic Fields with Reciprocity (NeurIPS 2025 · similarity 0.86)
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.
