Abstract
Document shadow removal is essential for enhancing the clarity of digitized documents. Preserving high-frequency details (e.g., text edges and lines) is critical in this process because shadows often obscure or distort fine structures. This paper proposes a matte vision transformer (MatteViT), a novel shadow removal framework that applies spatial and frequency-domain information to eliminate shadows while preserving fine-grained structural details. To effectively retain these details, we employ two preservation strategies. First, our method introduces a lightweight high-frequency amplification module (HFAM) that decomposes and adaptively amplifies high-frequency components. Second, we present a continuous luminance-based shadow matte, generated using a custom-built matte dataset and shadow matte generator, which provides precise spatial guidance from the earliest processing stage. These strategies enable the model to accurately identify fine-grained regions and restore them with high fidelity. Extensive experiments on public benchmarks (RDD and Kligler) demonstrate that MatteViT achieves state-of-the-art performance, providing a robust and practical solution for real-world document shadow removal. Furthermore, the proposed method better preserves text-level details in downstream tasks, such as optical character recognition, improving recognition performance over prior methods.
Abstract (translated)
文档阴影去除对于提升数字化文档的清晰度至关重要。保留高频细节（例如文本边缘和线条）在这个过程中非常关键，因为阴影往往会遮蔽或扭曲细小结构。本文提出了一种新颖的阴影去除框架——MatteViT（图层视觉变换器），它结合了空间信息与频域信息来消除阴影的同时保持精细的结构细节。为了有效保留这些细节，我们采用两种保护策略。首先，我们的方法引入了一个轻量级高频放大模块（HFAM），该模块可以分解并自适应地增强高频分量。其次，我们提出了一种基于连续亮度的图层生成技术，通过使用定制构建的数据集和阴影图层生成器来提供精确的空间引导信息，从最早的处理阶段开始便发挥作用。这些策略使模型能够准确识别精细区域，并以高保真度进行恢复。 在公共基准测试（RDD和Kligler）上的广泛实验表明，MatteViT达到了最先进的性能水平，为现实世界中的文档阴影去除提供了稳健且实用的解决方案。此外，所提出的方法在下游任务中更好地保留了文本级别的细节，例如光学字符识别，其识别准确率优于先前方法。
URL
https://arxiv.org/abs/2512.08789
