# Think Before You Restore: Risk-Aware Manchu Manuscript Restoration with Stroke-Guided Attention

arXiv: https://arxiv.org/abs/2609.36243 · HTML: https://mlsanity.com/paper/2609.36243/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 13 of its 25 references that are in the map.
- First posted on arXiv 28 Sep 2026.

## Builds on (most cited first, 13 of 13)

- [Restormer: Efficient Transformer for High-Resolution Image Restoration](https://mlsanity.com/paper/2111.09881/) (2021)
- [MUSIQ: Multi-scale Image Quality Transformer](https://mlsanity.com/paper/2108.05997/) (2021)
- [Exploring CLIP for Assessing the Look and Feel of Images](https://mlsanity.com/paper/2207.12396/) (2022)
- [Simple Baselines for Image Restoration](https://mlsanity.com/paper/2204.04676/) (2022)
- [Free-Form Image Inpainting with Gated Convolution](https://mlsanity.com/paper/1806.03589/) (2018)
- [MANIQA: Multi-dimension Attention Network for No-Reference Image Quality Assessment](https://mlsanity.com/paper/2204.08958/) (2022)
- [MAT: Mask-Aware Transformer for Large Hole Image Inpainting](https://mlsanity.com/paper/2203.15270/) (2022)
- [Gradient Magnitude Similarity Deviation: A Highly Efficient Perceptual Image Quality Index](https://mlsanity.com/paper/1308.3052/) (2013)
- [Harnessing Diffusion-Yielded Score Priors for Image Restoration](https://mlsanity.com/paper/2507.20590/) (2025)
- [Text Gestalt: Stroke-Aware Scene Text Image Super-Resolution](https://mlsanity.com/paper/2112.08171/) (2021)
- [DocRes: A Generalist Model Toward Unifying Document Image Restoration Tasks](https://mlsanity.com/paper/2405.04408/) (2024)
- [Predicting the Original Appearance of Damaged Historical Documents](https://mlsanity.com/paper/2412.11634/) (2024)
- [Text Image Inpainting via Global Structure-Guided Diffusion Models](https://mlsanity.com/paper/2401.14832/) (2024)

## Abstract (arXiv)

Full-page blind restoration of historical Manchu manuscripts is challenging due to scarce annotations, unknown degradation regions, and fragile connected strokes. Generic restoration models may improve visual quality but often modify intact content, leading to over-restoration. We propose SAGE-Restore (Stroke-Aware Gated rEstoration), a selective restoration framework that first assesses where restoration is needed and then uses this assessment to guide restoration candidate generation and pixel-level selection. Its encoder predicts patch-level repair probabilities from complementary appearance and stroke-structural cues to condition restoration candidate generation, while the corresponding repair logits are refined into a pixel-level soft gate that selectively controls where the restoration candidate is applied. We further introduce a fidelity-aware evaluation protocol that jointly measures degraded-region recovery, intact-content preservation, and their balance. SAGE-Restore achieves the highest R-Recovery (0.463) and RFS (0.626), while maintaining high U-Fidelity (0.968), demonstrating an effective balance between restoration and content preservation.