Research map / Vision-language and multimodal / Human and embodied modeling
DAN: A Segmentation-Free Document Attention Network for Handwritten Document Recognition
Denis Coquenet, Clément Chatelain, Thierry Paquet
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 |
|---|---|---|---|
| add synthetic document pre-generation (ablation) | DAN · READ 2016 (single-page) | CER (%): 81.05 -> 3.65 | Improved |
| add synthetic document pre-generation (ablation) | DAN · RIMES 2009 (single-page) | CER (%): 8.26 -> 5.72 | Improved |
| add curriculum on number of lines for synthetic data (ablation) | DAN · READ 2016 (double-page) | CER (%): 78.89 -> 4.50 | Improved |
| add cropping below last text line in curriculum (ablation) | DAN · READ 2016 (single-page) | CER (%): 100.00 -> 3.65 | Improved |
| add data augmentation (ablation) | DAN · RIMES 2009 (single-page) | CER (%): 7.08 -> 5.72 | Improved |
| add curriculum dropout (ablation) | DAN · READ 2016 (single-page) | CER (%): 3.65 -> 3.92 | Hurt |
| add curriculum dropout (ablation) | DAN · READ 2016 (double-page) | CER (%): 4.23 -> 4.50 | Improved |
| add errors in teacher forcing (ablation) | DAN · READ 2016 (double-page) | CER (%): 85.78 -> 4.50 | Improved |
| add layout token recognition (ablation) | DAN · RIMES 2009 (single-page) | CER (%): 5.30 -> 5.72 | Hurt |
| add layout token recognition (ablation) | DAN · READ 2016 (single-page) | CER (%): 4.60 -> 3.65 | Improved |
| add pre-training (ablation) | DAN · RIMES 2009 (single-page) | CER (%): 71.42 -> 5.72 | Improved |
| add pre-training (ablation) | DAN · READ 2016 (double-page) | CER (%): 5.84 -> 4.50 | Improved |
Closest papers by meaning
- Decoupled Attention Network for Text Recognition (AAAI 2020 · similarity 0.90)
- End-to-End Handwritten Paragraph Text Recognition Using a Vertical Attention Network (TPAMI 2022 · similarity 0.89)
- Handwritten Text Recognition: A Survey (TPAMI 2025 · similarity 0.87)
- An End-to-End TextSpotter With Explicit Alignment and Attention (CVPR 2018 · similarity 0.87)
- OrigamiNet: Weakly-Supervised, Segmentation-Free, One-Step, Full Page Text Recognition by learning to unfold (CVPR 2020 · similarity 0.86)
- MetaWriter: Personalized Handwritten Text Recognition Using Meta-Learned Prompt Tuning (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.
