Neural Enhancement of Analytical Appearance Models
This work addresses the trade-off between accuracy and efficiency in appearance modeling for computer graphics, offering a practical solution that combines the interpretability of analytical models with the expressiveness of neural networks.
The paper introduces neural enhancement, a framework that replaces key computational nodes in analytical reflectance models with small MLPs to improve accuracy while maintaining compactness and efficiency. The enhanced models achieve state-of-the-art results on fitting measured reflectance and bidirectional texture functions.
Traditional analytical reflectance models, while compact and interpretable, lack the capacity to accurately represent physical measurements. Recent neural models, which closely fit input data, are less generalizable and often more expensive to store and evaluate. To combine the strengths and overcome the limitations of these two classes of models, we present neural enhancement, a novel framework to boost an input analytical appearance model, by identifying and replacing its key computational nodes/operators with small-scale multi-layer perceptrons. This allows us to leverage the computational graph structure of the original model, while improving its expressiveness at a modest cost. To make the enhancement computationally tractable, we propose a hypercube-based search to automatically and efficiently identify the node(s) and/or operator(s) to be replaced towards maximal gain in a differentiable fashion. We enhance a number of common analytical BRDF models. The results are, at once accurate, compact and efficient, and compare favorably with state-of-the-art work on fitting measured reflectance and bidirectional texture functions. Finally, our models are fully compatible with any standard rasterization or ray-tracing pipeline.