Low-light image enhancement, particularly in cross-domain tasks such as mapping from the raw domain to the sRGB domain, remains a significant challenge. Many deep learning-based methods have been developed to address this issue and have shown promising results in recent years. However, single-stage methods, which attempt to unify the complex mapping across both domains, leading to limited denoising performance. In contrast, existing two-stage approaches typically overlook the characteristic of demosaicing within the Image Signal Processing (ISP) pipeline, leading to color distortions under varying lighting conditions, especially in low-light scenarios. To address these issues, we propose a novel Mamba-based method customized for low light RAW images, called RAWMamba, to effectively handle raw images with different CFAs. Furthermore, we introduce a Retinex Decomposition Module (RDM) grounded in Retinex prior, which decouples illumination from reflectance to facilitate more effective denoising and automatic non-linear exposure correction, reducing the effect of manual linear illumination enhancement. By bridging demosaicing and denoising, better enhancement for low light RAW images is achieved. Experimental evaluations conducted on public datasets SID and MCR demonstrate that our proposed RAWMamba achieves state-of-the-art performance on cross-domain mapping. The code is available at https://github.com/Cynicarlos/RetinexRawMamba.
30 August 2021 in IEEE Transactions on Neural Networks and Learning Systems
Face parsing aims to assign pixel-wise semantic labels to different facial components (e.g., hair, brows, and lips) in given face images. However, directly predicting pixel-level labels for each facial component over the whole face image would obtain limited accuracy, especially for tiny facial components. To address this problem, some recent works propose to first crop tiny patches from the whole face image and then predict masks for each facial component. However, such cropping-and-segmenting strategy consists of two independent stages, which cannot be jointly optimized. Besides, as one valuable piece of information for parsing the highly structured facial components, context cues are not elaborately explored by the existing works. To address these issues, we propose a component-level refinement network (CLRNet) for precisely segmenting out each facial component. Specifically, we introduce an attention mechanism to bridge the two independent stages together and form an end-to-end trainable pipeline for face parsing. Furthermore, we incorporate the global context information into the refining process for each cropped facial component patch, providing informative cues for accurate parsing. Extensive experiments are carried out on two benchmark datasets, LFW-PL and HELEN. The results demonstrate the superiority of the proposed CLRNet over other state-of-the-art methods, especially for tiny facial components.
Defocus deblurring is an important task in computer vision that aims to bring images back to clarity. Over recent years, both blind defocuse deblurring and non-blind defocuse deblurring methods have made great progress in the single image defocus deblurring task. However, most of existing methods lack direct enhancement of structural information, resulting in artifacts and structural distortion. To address this challenge, we propose a single image defocus deblurring method to improve the image quality of defocus deblurring based on structural information enhancement. Specifically, we introduce a spatial frequency interaction module to enhance the structure information and texture information of the image from both local and global perspectives. This enhancement is achieved through parallel processing of the features in the spatial and frequency domain. Moreover, we introduce a region information guidance module, which meticulously processes the clear and blurred regions independently during the learning process. This approach minimizes the adverse effects on the clear regions, effectively eliminates artifacts and distorted structures, and preserves the authentic structural and textural information of the image. To evaluate the superiority of the proposed method, we implement extensive experiments on three public benchmark datasets, DPDD, RealDOF and LFDOF. The experimental results demonstrate that our method achieves state-of-the-art performance.
24 August 2026 in Biomedical Signal Processing and Control