BRDF Measurement and Fitting Data for Three Rough Metallic Surfaces

This data publication provides image-based BRDF (bidirectional reflection distribution function) measurement results of three isotropic metallic samples with different surface roughness levels. The raw data is supplied as tab-separated text files containing incident and outgoing directions and the corresponding BRDF values, measured using a calibrated light stage at the University of Tübingen. The measurements represent BRDF values up to an unknown global scaling factor, which was compensated during model fitting. Evaluation and modeling were performed at the Physikalisch-Technische Bundesanstalt (PTB). A Cook–Torrance microfacet BRDF model with GGX normal distribution was fitted to the measurements. For each sample, 32 final parameter sets are provided, corresponding to all combinations of four final optimization stages, four weighting schemes, and two model variants, with and without a diffuse term. The dataset also includes RMS error statistics and visual comparisons between measurements and fitted model predictions. Since the single-GGX model can reproduce the measured BRDF profiles only to a limited extent, an additional double-GGX model was fitted to the data. The corresponding results are provided as supplementary material, together with a concise description of the adjusted model.