DIME-Net: A Dual-Illumination Adaptive Enhancement Network Based on Retinex and Mixture-of-Experts

2025-08-19

AI Summary

To address the challenge of diverse image degradations under complex illumination conditions (e.g., low-light, backlighting) and the poor generalizability of existing methods, this paper proposes a dual-illumination adaptive enhancement framework. Grounded in Retinex theory, the framework integrates illumination-aware cross-attention and sequential global attention modules to jointly model illumination conditions and suppress artifacts. A novel sparse gating mechanism dynamically routes inputs to multiple S-curve expert networks, enabling adaptive specialization for distinct degradation patterns. Furthermore, a Mixture-of-Experts (MoE) strategy is employed for end-to-end illumination estimation. Extensive experiments on both synthetic and real-world low-light/backlight datasets demonstrate state-of-the-art performance. Crucially, the framework achieves zero-shot cross-scenario generalization without retraining, significantly improving not only perceptual image quality but also downstream vision tasks—including object detection and semantic segmentation.

Technology Category

• Machine Learning: Mixture of Experts (MoE)
• Computer Vision: Diffusion Models for Vision
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Image degradation caused by complex lighting conditions such as low-light and backlit scenarios is commonly encountered in real-world environments, significantly affecting image quality and downstream vision tasks. Most existing methods focus on a single type of illumination degradation and lack the ability to handle diverse lighting conditions in a unified manner. To address this issue, we propose a dual-illumination enhancement framework called DIME-Net. The core of our method is a Mixture-of-Experts illumination estimator module, where a sparse gating mechanism adaptively selects suitable S-curve expert networks based on the illumination characteristics of the input image. By integrating Retinex theory, this module effectively performs enhancement tailored to both low-light and backlit images. To further correct illumination-induced artifacts and color distortions, we design a damage restoration module equipped with Illumination-Aware Cross Attention and Sequential-State Global Attention mechanisms. In addition, we construct a hybrid illumination dataset, MixBL, by integrating existing datasets, allowing our model to achieve robust illumination adaptability through a single training process. Experimental results show that DIME-Net achieves competitive performance on both synthetic and real-world low-light and backlit datasets without any retraining. These results demonstrate its generalization ability and potential for practical multimedia applications under diverse and complex illumination conditions.

Problem

• Enhances images degraded by low-light and backlit conditions
• Handles diverse lighting scenarios in unified framework
• Corrects illumination artifacts and color distortions adaptively

Innovation

• Mixture-of-Experts illumination estimator module
• Retinex theory integration for dual enhancement
• Illumination-Aware Cross Attention restoration mechanisms