🤖 AI Summary
Existing shadow removal methods often rely on hard-to-obtain shadow masks, limiting their generalizability. To address this, we propose ReHiT—the first mask-free single-image shadow removal framework grounded in Retinex theory. ReHiT employs a dual-branch CNN-Transformer architecture to separately model reflectance and illumination. We introduce the Illumination-Guided Histogram Transformer Block (IGHB), the first of its kind, which integrates Retinex-based decomposition principles with multi-scale semantic modeling to accurately capture complex, spatially varying shadows under non-uniform illumination. Additionally, residual dense feature learning is incorporated to enhance representational capacity. Evaluated on the NTIRE 2025 benchmark, ReHiT achieves state-of-the-art performance with the smallest parameter count and fastest inference speed, significantly improving practical deployability in real-world scenarios.
📝 Abstract
While deep learning methods have achieved notable progress in shadow removal, many existing approaches rely on shadow masks that are difficult to obtain, limiting their generalization to real-world scenes. In this work, we propose ReHiT, an efficient mask-free shadow removal framework based on a hybrid CNN-Transformer architecture guided by Retinex theory. We first introduce a dual-branch pipeline to separately model reflectance and illumination components, and each is restored by our developed Illumination-Guided Hybrid CNN-Transformer (IG-HCT) module. Second, besides the CNN-based blocks that are capable of learning residual dense features and performing multi-scale semantic fusion, multi-scale semantic fusion, we develop the Illumination-Guided Histogram Transformer Block (IGHB) to effectively handle non-uniform illumination and spatially complex shadows. Extensive experiments on several benchmark datasets validate the effectiveness of our approach over existing mask-free methods. Trained solely on the NTIRE 2025 Shadow Removal Challenge dataset, our solution delivers competitive results with one of the smallest parameter sizes and fastest inference speeds among top-ranked entries, highlighting its applicability for real-world applications with limited computational resources. The code is available at https://github.com/dongw22/oath.