# DR-Mamba: Automatic Inference-Time Domain Adaptation for Document Image Binarization via Sample-Conditioned Detail-Background Suppression

arXiv: https://arxiv.org/abs/2606.22625 · HTML: https://mlsanity.com/paper/2606.22625/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Mamba for image restoration, one of 10 directions in State-space sequence models (S4, Mamba).
- Builds on 16 of its 32 references that are in the map.
- First posted on arXiv 21 Jun 2026.

## Placed in

- State-space sequence models (S4, Mamba) / [Mamba for image restoration](https://mlsanity.com/ssm/mamba-for-image-restoration/)

## Builds on (most cited first, 16 of 16)

- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Mamba: Linear-Time Sequence Modeling with Selective State Spaces](https://mlsanity.com/paper/2312.00752/) (2023)
- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [Efficiently Modeling Long Sequences with Structured State Spaces](https://mlsanity.com/paper/2111.00396/) (2021)
- [Mixed Precision Training](https://mlsanity.com/paper/1710.03740/) (2017)
- [Vision Mamba: Efficient Visual Representation Learning with Bidirectional State Space Model](https://mlsanity.com/paper/2401.09417/) (2024)
- [VMamba: Visual State Space Model](https://mlsanity.com/paper/2401.10166/) (2024)
- [clDice -- A Novel Topology-Preserving Loss Function for Tubular Structure Segmentation](https://mlsanity.com/paper/2003.07311/) (2020)
- [Tversky loss function for image segmentation using 3D fully convolutional deep networks](https://mlsanity.com/paper/1706.05721/) (2017)
- [Boundary loss for highly unbalanced segmentation](https://mlsanity.com/paper/1812.07032/) (2018)
- [DocDiff: Document Enhancement via Residual Diffusion Models](https://mlsanity.com/paper/2305.03892/) (2023)
- [DocEnTr: An End-to-End Document Image Enhancement Transformer](https://mlsanity.com/paper/2201.10252/) (2022)
- [Document Image Binarization with Fully Convolutional Neural Networks](https://mlsanity.com/paper/1708.03276/) (2017)
- [NAF-DPM: A Nonlinear Activation-Free Diffusion Probabilistic Model for Document Enhancement](https://mlsanity.com/paper/2404.05669/) (2024)
- [DocBinFormer: A Two-Level Transformer Network for Effective Document Image Binarization](https://mlsanity.com/paper/2312.03568/) (2023)
- [MFE-GAN: Efficient GAN-based Framework for Document Image Enhancement and Binarization with Multi-scale Feature Extraction](https://mlsanity.com/paper/2512.14114/) (2025)

## Abstract (arXiv)

Degraded document image binarization is sensitive to domain shifts caused by paper aging, bleed-through, stains, shadows, and uneven illumination, and the foreground-background separation of recent learning-based methods can become unstable on unseen degradation domains. We propose DR-Mamba, a sample-conditioned detail-background suppression framework that performs automatic inference-time domain adaptation for document image binarization. Unlike test-time adaptation methods that require gradient updates or auxiliary data at inference, DR-Mamba adapts to each input document through input-dependent gates within a single forward pass, requiring no target-domain labels, no fine-tuning, and no test-time parameter updates. Instead of using Mamba-style selective scanning as a single generic feature path, DR-Mamba reinterprets it as fast-slow route modeling: a fast detail route captures local stroke structures, while a slow background route accumulates spatially persistent degradation responses. The two routes are integrated through an input-dependent subtractive gate that explicitly suppresses background interference rather than fusing features by addition or concatenation. We further add full-resolution detail-guided reconstruction and thin-stroke-aware supervision to recover fine strokes lost during downsampling. Evaluated under a leave-one-year-out protocol on DIBCO-style benchmarks, where each held-out year is treated as an unseen degradation domain, DR-Mamba shows that per-document, per-location subtractive suppression improves cross-domain robustness, with particularly strong performance on the most severely degraded held-out fold.