PolarScale: A Physics-Grounded Benchmark for Radiometrically Consistent RGB-to-Stokes Estimation

2026-10-06

AI Summary

This study addresses the physical inconsistency arising from the missing radiometric scale when inferring polarization from RGB images. We propose PolarScale, a benchmark that reformulates this unidentifiable radiometric scale into a dataset-conditioned semantic estimation task for the first time. Leveraging trichromatic full-Stokes measurement data, we conduct systematic evaluations using both restoration-based and generation-based backbone networks alongside multidimensional physical metrics. The optimal model achieves a scale error as low as 3.6% and a physical violation rate below 0.25%, substantially enhancing performance in diffuse reflection separation, material segmentation, and glare classification.

Abstract

Polarization imaging provides physical cues beyond intensity imaging but typically requires specialized hardware. Recent methods infer polarization from RGB-like inputs, yet predict only normalized Stokes components or relative descriptors, from which the radiometric scale needed for full Stokes reconstruction has been divided out. We introduce PolarScale, a benchmark that makes this scale an explicit prediction and evaluation target. Built on existing trichromatic full-Stokes measurements, PolarScale takes the per-scene normalized total-intensity image $s_0$ (a scene-referred linear image, not a consumer sRGB photograph) and asks models to predict normalized Stokes components, AoLP/DoLP/DoCP, and a per-scene scale. Because the scale is divided out of the input, it is not physically identifiable; PolarScale therefore evaluates dataset-conditioned semantic scale estimation against a constant-scale control, together with angular, self-consistency, and physical-bound metrics. Across seven restoration-based and generative backbones and three prediction strategies, the strongest restoration models estimate the scale with 3.6-4.3% mean relative error versus 5.7% for the constant control and violate physical bounds on fewer than 0.25% of pixels, whereas two generative baselines collapse to a near-zero scale; explicit descriptor supervision improves descriptor accuracy (23.66 vs. 18.88 dB PSNR for MAE). Predicted full-Stokes representations improve diffuse/specular separation, material segmentation, and glare classification, although in diffuse/specular separation the learned scale performs only on par with the constant control.

Problem

• polarization imaging
• Stokes estimation
• radiometric scale
• RGB-to-polarization
• benchmark

Innovation

• PolarScale benchmark
• Radiometric scale estimation
• Full-Stokes reconstruction
• Physics-grounded evaluation
• Descriptor supervision