Research map / Vision-language and multimodal / Human and embodied modeling
ShreddingNet: Coarse-to-Fine Restoration for Multi-Source Shredded Manuscripts
Haoyang Cui, Hao Jiang, Yadong Mu
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add fragment clustering (label propagation on coarse candidates) (ablation) | ShreddingNet · art 2192 (3 sources) | ARI (%): 53.44 -> 92.29 | Improved |
| add fragment clustering (ablation) | ShreddingNet · art 2192 (3 sources) | global assembly precision / recall (%): 97.94 -> 99.15 prec; 97.31 -> 97.59 rec | Improved |
| add fine-grained matching stage (ablation) | ShreddingNet · art 2192 (3 sources) | fine-grained precision (%): 20.77 -> 77.07 | Improved |
| add fine-grained matching stage (ablation) | ShreddingNet · art 2192 (3 sources) | fine-grained recall (%): 98.49 -> 78.40 | Hurt |
| use CNN-based score evaluation instead of rule-based scoring (ablation) | ShreddingNet · art 2192 (3 sources) | global assembly precision/recall (%): 96.43 -> 99.15 prec; 93.54 -> 97.59 rec | Improved |
| increase Top-K from small values up to K=20 (ablation) | ShreddingNet · art 2192 (3 sources) | global metrics | Improved |
| increase Top-K beyond 20 (ablation) | ShreddingNet · art 2192 (3 sources) | global precision/F1 | No clear effect |
| add degradations (stains, mold, contour defects) to test fragments | ShreddingNet · art 2192 (3 sources) | global recall (%): 97.59 -> 81.43 | Hurt |
Closest papers by meaning
- Towards Automated Chinese Ancient Character Restoration: A Diffusion-Based Method with a New Dataset (AAAI 2024 · similarity 0.85)
- Predicting the Original Appearance of Damaged Historical Documents (AAAI 2025 · similarity 0.85)
- ReFIR: Grounding Large Restoration Models with Retrieval Augmentation (NeurIPS 2024 · similarity 0.83)
- 3DPGS: 3D Probabilistic Graph Search for Archaeological Piece Grouping (AAAI 2025 · similarity 0.83)
- Descanning: From Scanned to the Original Images with a Color Correction Diffusion Model (AAAI 2024 · similarity 0.83)
- Self-Supervised Large Scale Point Cloud Completion for Archaeological Site Restoration (CVPR 2025 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
