IllumFlow: Illumination-Adaptive Low-Light Enhancement via Conditional Rectified Flow and Retinex Decomposition
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914135489380352 |
|---|---|
| author | Wei, Wenyang yang, Yang Jia, Xixi Feng, Xiangchu Wang, Weiwei Wang, Renzhen |
| author_facet | Wei, Wenyang yang, Yang Jia, Xixi Feng, Xiangchu Wang, Weiwei Wang, Renzhen |
| contents | We present IllumFlow, a novel framework that synergizes conditional Rectified Flow (CRF) with Retinex theory for low-light image enhancement (LLIE). Our model addresses low-light enhancement through separate optimization of illumination and reflectance components, effectively handling both lighting variations and noise. Specifically, we first decompose an input image into reflectance and illumination components following Retinex theory. To model the wide dynamic range of illumination variations in low-light images, we propose a conditional rectified flow framework that represents illumination changes as a continuous flow field. While complex noise primarily resides in the reflectance component, we introduce a denoising network, enhanced by flow-derived data augmentation, to remove reflectance noise and chromatic aberration while preserving color fidelity. IllumFlow enables precise illumination adaptation across lighting conditions while naturally supporting customizable brightness enhancement. Extensive experiments on low-light enhancement and exposure correction demonstrate superior quantitative and qualitative performance over existing methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_02411 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | IllumFlow: Illumination-Adaptive Low-Light Enhancement via Conditional Rectified Flow and Retinex Decomposition Wei, Wenyang yang, Yang Jia, Xixi Feng, Xiangchu Wang, Weiwei Wang, Renzhen Computer Vision and Pattern Recognition We present IllumFlow, a novel framework that synergizes conditional Rectified Flow (CRF) with Retinex theory for low-light image enhancement (LLIE). Our model addresses low-light enhancement through separate optimization of illumination and reflectance components, effectively handling both lighting variations and noise. Specifically, we first decompose an input image into reflectance and illumination components following Retinex theory. To model the wide dynamic range of illumination variations in low-light images, we propose a conditional rectified flow framework that represents illumination changes as a continuous flow field. While complex noise primarily resides in the reflectance component, we introduce a denoising network, enhanced by flow-derived data augmentation, to remove reflectance noise and chromatic aberration while preserving color fidelity. IllumFlow enables precise illumination adaptation across lighting conditions while naturally supporting customizable brightness enhancement. Extensive experiments on low-light enhancement and exposure correction demonstrate superior quantitative and qualitative performance over existing methods. |
| title | IllumFlow: Illumination-Adaptive Low-Light Enhancement via Conditional Rectified Flow and Retinex Decomposition |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2511.02411 |