基于物理规律的双分支模型，统一修复多种超高清图像退化问题
RetinexDualV2: Physically-Grounded Dual Retinex for Generalized UHD Image Restoration
- 用物理先验引导图像分解，显式建模雨滴、暗通道等退化特征
- 在两个挑战赛中分别获第4和第5名，支持多种复杂退化场景统一处理
- 适合需要通用化图像恢复的科研与工业应用
我们提出RetinexDualV2，一种统一的、物理基础驱动的双分支框架，用于应对多样化的超高清（UHD）图像修复任务。与通用模型不同，该方法引入任务特定的物理先验模块（TS-PGM），提取雨滴掩码、暗通道等退化感知先验，并通过新型物理条件多头自注意力（PC-MSA）机制，显式指导Retinex分解网络进行反射与光照修正。这种物理约束使单一架构可无缝处理多种复杂退化，无需针对每项任务调整结构。RetinexDualV2在NTIRE 2026日夜间雨滴去除挑战赛中位列第4，在联合噪声与低光增强（JNLLIE）挑战赛中位列第5。大量实验验证了其领先性能与高效性。代码已公开于https://github.com/ErrorLogic1211/RetinexDual/tree/master/RetinexDualV2。
原文摘要 · Abstract (English)
We propose RetinexDualV2, a unified, physically grounded dual-branch framework for diverse Ultra-High-Definition (UHD) image restoration. Unlike generic models, our method employs a Task-Specific Physical Grounding Module (TS-PGM) to extract degradation-aware priors (e.g., rain masks and dark channels). These explicitly guide a Retinex decomposition network via a novel Physical-Conditioned Multi-head Self-Attention (PC-MSA) mechanism, enabling robust reflection and illumination correction. This physical conditioning allows a single architecture to handle various complex degradations seamlessly, without task-specific structural modifications. RetinexDualV2 demonstrates exceptional generalizability, securing 4th place in the NTIRE 2026 Day and Night Raindrop Removal Challenge and 5th place in the Joint Noise Low-light Enhancement (JNLLIE) Challenge. Extensive experiments confirm the state-of-the-art performance and efficiency of our physically motivated approach. Code is available at https://github.com/ErrorLogic1211/RetinexDual/tree/master/RetinexDualV2
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。