← Back to all news
arXiv Computer Vision By Justine Giroux, Jack Oliver Hilliard, Yannick Hold-Geoffroy, Javier Vazquez-Corral, Jean-Fran\c{c}ois Lalonde

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

Read the original on arXiv Computer Vision →

The paper introduces a benchmark called Shedding Light to evaluate how well generative image models understand and reproduce lighting. The benchmark tests models by asking them to inpaint a simple object, called a light probe, into real photographs and then compares the generated probe to the ground truth to assess lighting direction, colour, and radiance. The authors provide a scalable protocol and open-source code and data for systematic assessment of photometric accuracy in future models.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computer Vision.

Related stories

arXiv Computer Vision
Oct 26

Do Image Editing Models Understand Lighting?

arXiv:2606.26738v3 Announce Type: replace Abstract: While recent advancements in generative image editing models have achieved stunning visual fidelity, it remains an open question whether these syst...

By Tim K\"uchler, Johann-Friedrich Feiden, Matthias Nie{\ss}ner, Carsten Rother
benchmarks
More like this →
arXiv AI
Sep 10

WildRelight: A Real-World Benchmark and Physics-Guided Adaptation for Single-Image Relighting

WildRelight is the first in-the-wild dataset designed to evaluate single-image relighting models, featuring high-resolution outdoor scenes captured under strictly aligned, temporally varying natural illuminations paired with high-dynamic-range environment maps. The benchmark demonstrates that state-of-the-art models trained on synthetic data suffer severe domain shifts when applied to real-world imagery. Leveraging the dataset’s temporal structure, the authors introduce a physics-guided inference framework combining Diffusion Posterior Sampling with Temporal Sampling-Aware Test-Time Adaptation, enabling synthetic models to self-supervise and align with real-world statistics on-the-fly.

By Lezhong Wang, Mehmet Onurcan Kaya, Siavash Bigdeli, Jeppe Revall Frisvad
diffusionroboticsfine-tuningbenchmarks
More like this →
Hugging Face Trending Papers
Oct 5

Casual Flash Lighting for Gaussian Splat Inverse Rendering

The paper introduces a method that combines casual indoor photographs taken with and without flash to recover geometry, materials, and lighting. By using flash residuals to constrain albedo and BRDF and static lighting to capture specular highlights, the approach employs a hash‑encoded MLP anchored to a 2DGS depth map for view‑consistent material decomposition. Experiments on synthetic and real scenes show improved diffuse color, albedo, roughness, and relighting performance, achieving a 4.17 dB PSNR gain over the best baseline.

More like this →