Learning Sparse BRDF Measurement Samples from Image

TL;DR

This paper introduces a method for selecting the most informative sparse BRDF measurements to efficiently reconstruct material appearance, leveraging a learned reflectance prior and differentiable rendering.

Contribution

It proposes a novel sampler that optimizes measurement locations using a fixed BRDF reconstructor and differentiable rendering, improving low-budget reconstruction quality.

Findings

The sampler outperforms neural baselines at 8 and 16 measurements.

PCA-based methods remain competitive at larger measurement budgets.

Analysis of image-space supervision and latent fitting effects.

Abstract

Accurate BRDF acquisition is important for realistic rendering, but dense gonioreflectometer measurements are slow and expensive. We study how to select a small number of BRDF measurements that are most useful for reconstructing material appearance under a learned reflectance prior. Our method combines a set encoder for sparse coordinate-value observations, a pretrained hypernetwork-based BRDF reconstructor, and a differentiable renderer. During sampler training, the reconstructor is kept fixed and gradients from BRDF-space and rendered-image losses are used to optimize measurement locations. This separates sample selection from prior fitting and encourages the sampler to choose directions that are informative under the learned material distribution. Experiments on the MERL dataset show that the proposed sampler improves low-budget reconstruction quality at 8 and 16 measurements compared with neural reconstruction baselines, while PCA-based methods remain strong at larger budgets. We further analyze the effect of image-space supervision, co-optimization, and image-only latent fitting for unseen materials.