Research map / Image generation / Image restoration
SyntheticDoc: A Large Synthetic Dataset for Document Unwarping and Illumination Correction
Daniel Woortmann, Tanguy Magne, Olga Sorkine-Hornung
Read the paperCheck your idea against 137,217 papers
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 |
|---|---|---|---|
| train on SyntheticDoc 1M instead of 100k samples (ablation) | UVDoc-style baseline · DocUNet benchmark | MS-SSIM: 0.532 -> 0.558 | Improved |
| train on 1M instead of 500k samples (ablation) | UVDoc-style baseline · DocUNet benchmark | AD: 0.278 -> 0.252 | Improved |
| increase input/output resolution to 1024x1440 / 121x177 (ablation) | UVDoc-style baseline · DocUNet benchmark | AD: 0.252 -> 0.219 | Improved |
| increase input/output resolution to 1024x1440 / 121x177 (ablation) | UVDoc-style baseline · DocUNet benchmark | CER / ED: CER 0.161 -> 0.164; ED 628 -> 651 | No clear effect |
| train lightweight baseline on SyntheticDoc instead of Doc3D+UVDoc-based SOTA | UVDoc-style baseline · DocUNet benchmark | MS-SSIM: 0.544 (UVDoc) -> 0.558 | Improved |
| use SyntheticDoc-trained baseline vs DocTr for joint unwarp+illumination | UVDoc-style baseline · DocUNet benchmark (joint) | CER: 0.126 (DocTr) -> 0.149 | Hurt |
| use SyntheticDoc-trained baseline vs DocRES for joint task | UVDoc-style baseline · DocUNet benchmark (joint) | PSNR: 11.68 -> 12.68 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- DocRES — DocRes: A Generalist Model Toward Unifying Document Image Restoration Tasks (CVPR 2024) · better on 3 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Descanning: From Scanned to the Original Images with a Color Correction Diffusion Model (AAAI 2024 · similarity 0.85)
- DocRes: A Generalist Model Toward Unifying Document Image Restoration Tasks (CVPR 2024 · similarity 0.85)
- Scaling Laws of Synthetic Images for Model Training ... for Now (CVPR 2024 · similarity 0.85)
- The Unmet Promise of Synthetic Training Images: Using Retrieved Real Images Performs Better (NeurIPS 2024 · similarity 0.85)
- DocNLC: A Document Image Enhancement Framework with Normalized and Latent Contrastive Representation for Multiple Degradations (AAAI 2024 · similarity 0.84)
- CO-SPY: Combining Semantic and Pixel Features to Detect Synthetic Images by AI (CVPR 2025 · similarity 0.84)
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.
