SOURCE-LINKED INTELLIGENCE

Shedding Light: A Benchmark for Evaluating Lighting Understanding in Generative Image Models

arXiv · AI, language, vision and robotics · article · Sep 9, 2026 · UTC

Accurate modelling of illumination is central to realistic image synthesis and scene understanding. Yet, there is little exploration into whether image generative models are good at this task or whether physical plausibility remains a key challenge for them. Clearly, significant progress has been made in realistic image synthesis, but do models truly understand lighting in a physically accurate manner? To answer this question, this work proposes a benchmark to assess the lighting understanding and harmonisation capabilities of generative models. Our key insight is that evaluating lighting unde

Read original source ↗ Open in workspace

recordType

paper

region

Global