Differentiable Acoustic Radiance Transfer
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866911596929875968 |
|---|---|
| author | Lee, Sungho Scerbo, Matteo Han, Seungu Choi, Min Jun Lee, Kyogu De Sena, Enzo |
| author_facet | Lee, Sungho Scerbo, Matteo Han, Seungu Choi, Min Jun Lee, Kyogu De Sena, Enzo |
| contents | Geometric acoustics is an efficient framework for room acoustics modeling, governed by the canonical time-dependent rendering equation. Acoustic radiance transfer (ART) solves the equation by discretization, modeling time- and direction-dependent energy exchange between surface patches with flexible material properties. We introduce DART, an efficient, differentiable implementation of ART that enables gradient-based optimization of material properties. We evaluate DART on a simpler variant of acoustic field learning that aims to predict energy responses for novel source-receiver configurations. Experimental results demonstrate that DART generalizes better under sparse measurement scenarios than existing signal processing and neural network baselines, while maintaining simplicity and full interpretability. We open-source our implementation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_15946 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Differentiable Acoustic Radiance Transfer Lee, Sungho Scerbo, Matteo Han, Seungu Choi, Min Jun Lee, Kyogu De Sena, Enzo Sound Audio and Speech Processing Signal Processing Geometric acoustics is an efficient framework for room acoustics modeling, governed by the canonical time-dependent rendering equation. Acoustic radiance transfer (ART) solves the equation by discretization, modeling time- and direction-dependent energy exchange between surface patches with flexible material properties. We introduce DART, an efficient, differentiable implementation of ART that enables gradient-based optimization of material properties. We evaluate DART on a simpler variant of acoustic field learning that aims to predict energy responses for novel source-receiver configurations. Experimental results demonstrate that DART generalizes better under sparse measurement scenarios than existing signal processing and neural network baselines, while maintaining simplicity and full interpretability. We open-source our implementation. |
| title | Differentiable Acoustic Radiance Transfer |
| topic | Sound Audio and Speech Processing Signal Processing |
| url | https://arxiv.org/abs/2509.15946 |