Problem
Research questions and friction points this paper is trying to address.
Manchu manuscript restoration
blind restoration
over-restoration
stroke preservation
Think Before You Restore: Risk-Aware Manchu Manuscript Restoration with Stroke-Guided Attention
🤖 AI Summary
This study addresses the challenge of over-restoration in blind restoration of Manchu ancient documents, caused by scarce annotations, unknown degradation regions, and fragile strokes. To this end, we propose SAGE-Restore, a selective restoration framework that pioneers patch-level restoration probability prediction based on stroke structural cues, transforming it into pixel-level soft gating to precisely control the restoration scope. By integrating dual encoders for appearance and structural features with a conditional candidate generation mechanism, the framework implements an "evaluate-then-guide" selective restoration paradigm. Experimental results demonstrate that SAGE-Restore achieves state-of-the-art performance on R-Recovery (0.463) and RFS (0.626) metrics while maintaining high-fidelity preservation with a U-Fidelity score of 0.968, effectively balancing precise restoration with the retention of original document integrity.
📝 Abstract
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.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Manchu Manuscript Restoration
Stroke-Guided Attention
Selective Restoration
Pixel-level Soft Gate
Fidelity-aware Evaluation
🔎 Similar Papers
No similar papers found.
