Light Up Your Face: A Physically Consistent Dataset and Diffusion Model for Face Fill-Light Enhancement

Jue Gong, Zihan Zhou, Jingkai Wang, Xiaohong Liu, Yulun Zhang, Xiaokang Yang
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:36045-36055, 2026.

Abstract

Face fill-light enhancement (FFE) brightens underexposed faces by adding virtual fill light while keeping the original scene illumination and background unchanged. Most face relighting methods aim to reshape overall lighting, which can suppress the input illumination or modify the entire scene, leading to foreground–background inconsistency and mismatching practical FFE needs. To support scalable learning, we introduce LightYourFace-160K (LYF-160K), a large-scale paired dataset built with a physically consistent renderer that injects a disk-shaped area fill light controlled by six disentangled factors, producing 160K before-and-after pairs. We first pretrain a physics-aware lighting prompt (PALP) that embeds the 6D parameters into conditioning tokens, using an auxiliary planar-light reconstruction objective. Building on a pretrained diffusion backbone, we then train a fill-light diffusion (FiLitDiff), an efficient one-step model conditioned on physically grounded lighting codes, enabling controllable and high-fidelity fill lighting at low computational cost. Experiments on held-out paired sets demonstrate strong perceptual quality and competitive full-reference metrics, while better preserving background illumination. The dataset and model will be at https://github.com/gobunu/Light-Up-Your-Face.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-gong26j, title = {Light Up Your Face: A Physically Consistent Dataset and Diffusion Model for Face Fill-Light Enhancement}, author = {Gong, Jue and Zhou, Zihan and Wang, Jingkai and Liu, Xiaohong and Zhang, Yulun and Yang, Xiaokang}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {36045--36055}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/gong26j/gong26j.pdf}, url = {https://proceedings.mlr.press/v306/gong26j.html}, abstract = {Face fill-light enhancement (FFE) brightens underexposed faces by adding virtual fill light while keeping the original scene illumination and background unchanged. Most face relighting methods aim to reshape overall lighting, which can suppress the input illumination or modify the entire scene, leading to foreground–background inconsistency and mismatching practical FFE needs. To support scalable learning, we introduce LightYourFace-160K (LYF-160K), a large-scale paired dataset built with a physically consistent renderer that injects a disk-shaped area fill light controlled by six disentangled factors, producing 160K before-and-after pairs. We first pretrain a physics-aware lighting prompt (PALP) that embeds the 6D parameters into conditioning tokens, using an auxiliary planar-light reconstruction objective. Building on a pretrained diffusion backbone, we then train a fill-light diffusion (FiLitDiff), an efficient one-step model conditioned on physically grounded lighting codes, enabling controllable and high-fidelity fill lighting at low computational cost. Experiments on held-out paired sets demonstrate strong perceptual quality and competitive full-reference metrics, while better preserving background illumination. The dataset and model will be at https://github.com/gobunu/Light-Up-Your-Face.} }
Endnote
%0 Conference Paper %T Light Up Your Face: A Physically Consistent Dataset and Diffusion Model for Face Fill-Light Enhancement %A Jue Gong %A Zihan Zhou %A Jingkai Wang %A Xiaohong Liu %A Yulun Zhang %A Xiaokang Yang %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-gong26j %I PMLR %P 36045--36055 %U https://proceedings.mlr.press/v306/gong26j.html %V 306 %X Face fill-light enhancement (FFE) brightens underexposed faces by adding virtual fill light while keeping the original scene illumination and background unchanged. Most face relighting methods aim to reshape overall lighting, which can suppress the input illumination or modify the entire scene, leading to foreground–background inconsistency and mismatching practical FFE needs. To support scalable learning, we introduce LightYourFace-160K (LYF-160K), a large-scale paired dataset built with a physically consistent renderer that injects a disk-shaped area fill light controlled by six disentangled factors, producing 160K before-and-after pairs. We first pretrain a physics-aware lighting prompt (PALP) that embeds the 6D parameters into conditioning tokens, using an auxiliary planar-light reconstruction objective. Building on a pretrained diffusion backbone, we then train a fill-light diffusion (FiLitDiff), an efficient one-step model conditioned on physically grounded lighting codes, enabling controllable and high-fidelity fill lighting at low computational cost. Experiments on held-out paired sets demonstrate strong perceptual quality and competitive full-reference metrics, while better preserving background illumination. The dataset and model will be at https://github.com/gobunu/Light-Up-Your-Face.
APA
Gong, J., Zhou, Z., Wang, J., Liu, X., Zhang, Y. & Yang, X.. (2026). Light Up Your Face: A Physically Consistent Dataset and Diffusion Model for Face Fill-Light Enhancement. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:36045-36055 Available from https://proceedings.mlr.press/v306/gong26j.html.

Related Material