This work recast color restoration as an independent sub-problem and decouple it from brightness enhancement, realizing it as a general-purpose post-processing module built on retrieval-augmented generation (RAG).
Abstract
Recent low-light image enhancement (LLIE) methods have driven brightness and structural fidelity close to that of normally-exposed images, yet their outputs still exhibit systematic color shifts such as greenish skies, yellowish faces, and warm-tinted white objects. We attribute this to end-to-end LLIE training coupling brightness, structure, and color within a single network, leaving the color channels weakly supervised. We recast color restoration as an independent sub-problem and decouple it from brightness enhancement, realizing it as a general-purpose post-processing module built on retrieval-augmented generation (RAG). Rather than relying solely on parametric color priors learned during training, the module dynamically retrieves a reference image from an external high-quality color knowledge base and injects its color distribution into a color-restoration network to correct residual bias. The design has three components: (i) a dual-index FAISS retriever built on intermediate VGG19 features, capturing textural and structural similarity through global mean and variance statistics; (ii) GlobalSPHistAdaIN, which reduces the reference spatial-preserving color histogram to a global color vector and modulates network features via adaptive instance normalization, removing dependence on pixel-level correspondence; and (iii) a residual formulation that predicts a color correction over the front-end output. Across LOLv1, LOLv2-Real, and LOLv2-Synthetic, the module consistently improves color-specific metrics, and it remains effective when the front end is swapped among CPGA-Net++, LLFormer, FLIGHTNet, and IAT, confirming cross-front-end generality. Ablations show that a VGG19 dual index outperforms CLIP-based retrieval, indicating that color restoration depends on textural and structural similarity rather than high-level semantics.
SPACE introduces a Depth-Adaptive HVI Transformation to decouple luminance and chrominance under depth guidance, effectively suppressing color-space noise and a Depth-Manifold Modulated Attention mechanism constrains feature interactions within a learned depth manifold, ensuring structural coherence during enhancement.
Yue Zhang, Zhi-Liang Wu, Yuxuan Hou et al.· Proceedings of the Thirty-Fi...· 0 citations
Image enhancement in low-light conditions is a challenging problem within the field of computer vision, since underexposed images generally lead to poor visibility, low contrast, noise amplification, and color distortion. Recent deep learning approaches have indeed shown promising performance, yet most of them adopt computation-heavy architectures and do not treat the luminance enhancement and color restoration separately; as a result cause unnatural restorations. To this end, in this paper, we present an efficient low-light image enhancement framework based on the Deep White-Balance (DWB) with Dark Channel Prior guidance in the YCbCr color space. The proposed framework separates luminance and chrominance components, making it easier to manage brightness enhancement and color restoration separately. This method aims to generate visually consistent enhanced images while preserving color fidelity and avoiding common enhancement artifacts. We evaluate the performance of the proposed method on reference benchmark datasets (LOL, LOLv2-Synthetic, and LIME) as well as a no-reference benchmark dataset (DICM). The experimental results demonstrate that, although state-of-the-art deep learning methods achieve higher numerical scores, qualitatively, the framework produces enhanced images with consistent contrast and illumination, while retaining color fidelity, all requiring less computational resources. Moreover, it does not require any further training or fine-tuning since the proposed approach is based on a pretrained Deep White Balance model and only uses inference. Experimental results show that the proposed method provides a convincing quality-efficiency compromise for low-light image enhancement.
S. J. Shahbaz, H. G. Daway, Ahlam M. Kadhim· Journal of Intelligent &...· 0 citations
Low-light imaging often introduces color bias caused by the low signal-to-noise ratio and the image formation process. Although recent low-light image enhancement methods have achieved strong brightness recovery, faithful color restoration remains challenging, manifesting as overall color bias together with local under- and over-saturation. To address this issue, we propose CAGE, a cylindrical color correction framework with adaptive color debiasing and gamut-harmonized saturation rectification for color-faithful low-light image enhancement. We first introduce AdaLAB, a cylindrical adaptive LAB color space that provides a decoupled and image-specific basis for uniform color correction. Building on this color space, we further develop AdaCCT, an adaptive cylindrical color transform with forward and inverse transforms for the conversion between RGB and AdaLAB color space, as well as necessary color debiasing and saturation rectification. The forward transform suppresses embedded color bias before backbone enhancement by reorganizing the chromatic distribution through chromatic-plane shifting and scaling, while the inverse transform achieves faithful saturation rectification through out-of-gamut lightness compensation. Extensive experiments on multiple benchmarks show that CAGE achieves more faithful color restoration, specifically reduces color bias and saturation abnormality, and delivers better overall visual quality across different low-light enhancement backbones. The code is available at https://yangzhichen763.github.io/CAGE/.
Zhichen Yang, Rui Xu, Yu-Zhen Niu et al.· 0 citations
SFH-Net is proposed, an horizontal-vertical-intensity (HVI)-guided luminance-chrominance collaborative enhancement framework that achieves a better trade-off among reconstruction accuracy, structural fidelity, and parameter compactness.
Zhanqiang Huo, Hui-Jie Zhang, Yingxu Qiao et al.· Engineering Research Express· 0 citations
This work proposes Structure-Anchored Rectified Flow (SA-RF), which maintains correspondence through separate chromaticity/intensity stems, a scale-matched condition pyramid, and HybridAda, and introduces BC-IHV, a learnable Box--Cox polar color space whose analytically invertible intensity mapping controls the inverse-gradient dynamic range through a single exponent.
Yihao Ai, Zheng Chen, Yuanhao Cai et al.· 0 citations
A Hybrid Prior Enhanced Decomposition (HPED) model is proposed, a training-free, model-driven framework that incorporates structural and luminance priors into a multi-stage enhancement pipeline, outperforming all competing methods and suggesting potential applicability in downstream machine perception tasks such as detection and tracking.
Jie Li, Cheng Wang, Xiang-Yu Li et al.· Remote Sensing· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
