Guardado en:
Detalles Bibliográficos
Autores principales: Liu, Hanzhou, Jiang, Peng, Huang, Jia, Lu, Mi
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:https://arxiv.org/abs/2511.09818
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866910147946741760
author Liu, Hanzhou
Jiang, Peng
Huang, Jia
Lu, Mi
author_facet Liu, Hanzhou
Jiang, Peng
Huang, Jia
Lu, Mi
contents Restoring 3D scenes with low-light conditions is challenging, and most existing methods depend on precomputed camera poses and scene-specific optimization, which greatly restricts their application to real-world scenarios. To overcome these limitations, we propose Lumos3D, a pose-free single-forward framework for 3D low-light scene restoration. First, we develop a cross-illumination distillation scheme, where a frozen teacher network takes normal-light ground truth images as input to distill accurate geometric information to the student model. Second, we define a Lumos loss to improve the restoration quality of the reconstructed 3D Gaussian space. Trained on a single dataset, Lumos3D performs inference in a purely feed-forward manner, directly restoring illumination and structure from unposed, low-light multi-view images without any per-scene training or optimization. Experiments on real-world datasets demonstrate that Lumos3D achieves competitive restoration results compared to scene-specific methods. Our codes will be released soon.
format Preprint
id arxiv_https___arxiv_org_abs_2511_09818
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Lumos3D: A Single-Forward Framework for Low-Light 3D Scene Restoration
Liu, Hanzhou
Jiang, Peng
Huang, Jia
Lu, Mi
Computer Vision and Pattern Recognition
Restoring 3D scenes with low-light conditions is challenging, and most existing methods depend on precomputed camera poses and scene-specific optimization, which greatly restricts their application to real-world scenarios. To overcome these limitations, we propose Lumos3D, a pose-free single-forward framework for 3D low-light scene restoration. First, we develop a cross-illumination distillation scheme, where a frozen teacher network takes normal-light ground truth images as input to distill accurate geometric information to the student model. Second, we define a Lumos loss to improve the restoration quality of the reconstructed 3D Gaussian space. Trained on a single dataset, Lumos3D performs inference in a purely feed-forward manner, directly restoring illumination and structure from unposed, low-light multi-view images without any per-scene training or optimization. Experiments on real-world datasets demonstrate that Lumos3D achieves competitive restoration results compared to scene-specific methods. Our codes will be released soon.
title Lumos3D: A Single-Forward Framework for Low-Light 3D Scene Restoration
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.09818