Differentiable Acoustic Radiance Transfer
For acoustics researchers and engineers, DART provides an interpretable, differentiable framework for optimizing room acoustics simulations, though the improvement is incremental over existing methods.
DART introduces a differentiable acoustic radiance transfer method for room acoustics, enabling gradient-based optimization of material properties. It outperforms signal processing and neural network baselines in predicting energy responses under sparse measurements, with better generalization.
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.