DRWKV: Focusing on Object Edges for Low-Light Image Enhancement

2025-07-24

AI Summary

To address edge discontinuity and detail loss under extreme low-light conditions, this paper proposes an illumination-edge decoupled enhancement framework. The method introduces three key innovations: (1) a global edge-aware Retinex model that explicitly decouples illumination and edge components; (2) a spiral-scan Evolving WKV attention mechanism to efficiently capture long-range edge dependencies; and (3) a Bi-Spectral Alignment Block (Bi-SAB) jointly optimizing luminance and chrominance consistency in the frequency domain, coupled with an MS2-Loss for multi-scale structural fidelity. Evaluated on five mainstream low-light benchmarks, the approach achieves state-of-the-art performance—significantly improving PSNR and SSIM while reducing NIQE by up to 18.7%. It incurs low computational overhead and notably enhances downstream low-light multi-object tracking accuracy, demonstrating strong generalization capability.

Technology Category

• Computer Vision: Low Level & Physics-based Vision
• Machine Learning: Learning on the Edge & Model Compression
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

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

Abstract

Low-light image enhancement remains a challenging task, particularly in preserving object edge continuity and fine structural details under extreme illumination degradation. In this paper, we propose a novel model, DRWKV (Detailed Receptance Weighted Key Value), which integrates our proposed Global Edge Retinex (GER) theory, enabling effective decoupling of illumination and edge structures for enhanced edge fidelity. Secondly, we introduce Evolving WKV Attention, a spiral-scanning mechanism that captures spatial edge continuity and models irregular structures more effectively. Thirdly, we design the Bilateral Spectrum Aligner (Bi-SAB) and a tailored MS2-Loss to jointly align luminance and chrominance features, improving visual naturalness and mitigating artifacts. Extensive experiments on five LLIE benchmarks demonstrate that DRWKV achieves leading performance in PSNR, SSIM, and NIQE while maintaining low computational complexity. Furthermore, DRWKV enhances downstream performance in low-light multi-object tracking tasks, validating its generalization capabilities.

Problem

• Enhancing low-light images while preserving object edges
• Decoupling illumination and edge structures for fidelity
• Improving visual naturalness and reducing artifacts

Innovation

• Integrates Global Edge Retinex for edge fidelity
• Uses Evolving WKV Attention for spatial continuity
• Aligns features with Bilateral Spectrum Aligner