Residual Quotient Learning for Zero-Reference Low-Light Image Enhancement
IEEE Transactions on Image Processing · 2025 · 34 · 365-378
DOI: 10.1109/tip.2024.3519997 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Image Processing, 2025. Verfasst von Chao Xie und 7 weiteren Autor:innen. Die Arbeit wurde bisher 20-mal zitiert (Crossref) und nennt 65 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Natural Science Foundation of the Anhui Higher Education Institutions of China, State Visiting Scholar Program of China Scholarship Council.
Abstract (Originalsprache)
Recently, neural networks have become the dominant approach to low-light image enhancement (LLIE), with at least one-third of them adopting a Retinex-related architecture. However, through in-depth analysis, we contend that this most widely accepted LLIE structure is suboptimal, particularly when addressing the non-uniform illumination commonly observed in natural images. In this paper, we present a novel variant learning framework, termed residual quotient learning, to substantially alleviate this issue. Instead of following the existing Retinex-related decomposition-enhancement-reconstruction process, our basic idea is to explicitly reformulate the light enhancement task as adaptively predicting the latent quotient with reference to the original low-light input using a residual learning fashion. By leveraging the proposed residual quotient learning, we develop a lightweight yet effective network called ResQ-Net. This network features enhanced non-uniform illumination modeling capabilities, making it more suitable for real-world LLIE tasks. Moreover, due to its well-designed structure and reference-free loss function, ResQ-Net is flexible in training as it allows for zero-reference optimization, which further enhances the generalization and adaptability of our entire framework. Extensive experiments on various benchmark datasets demonstrate the merits and effectiveness of the proposed residual quotient learning, and our trained ResQ-Net outperforms state-of-the-art methods both qualitatively and quantitatively. Furthermore, a practical application in dark face detection is explored, and the preliminary results confirm the potential and feasibility of our method in real-world scenarios.
Förderung
- National Natural Science Foundation of China
- Natural Science Foundation of the Anhui Higher Education Institutions of China
- State Visiting Scholar Program of China Scholarship Council
- National College Student Practice and Innovation Training Program
Zitieren
APA
Xie, C., Fei, L., Tao, H., Hu, Y., Zhou, W., Hoe, J. T., Hu, W., & Tan, Y. P. (2025). Residual Quotient Learning for Zero-Reference Low-Light Image Enhancement. IEEE Transactions on Image Processing, 34, 365-378. https://doi.org/10.1109/tip.2024.3519997
Vancouver
Xie C, Fei L, Tao H, Hu Y, Zhou W, Hoe JT, et al. Residual Quotient Learning for Zero-Reference Low-Light Image Enhancement. IEEE Transactions on Image Processing. 2025;34:365-378. doi:10.1109/tip.2024.3519997
BibTeX
@article{Xie2025,
author = {Xie, Chao and Fei, Linfeng and Tao, Huanjie and Hu, Yaocong and Zhou, Wei and Hoe, Jiun Tian and Hu, Weipeng and Tan, Yap-Peng},
title = {Residual Quotient Learning for Zero-Reference Low-Light Image Enhancement},
journal = {IEEE Transactions on Image Processing},
year = {2025},
volume = {34},
pages = {365--378},
doi = {10.1109/tip.2024.3519997}
}
EUVIMED