Synthetic Data Augmentation for Table Detection: Re-evaluating TableNet's Performance with Automatically Generated Document Images
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866912436351664128 |
|---|---|
| author | Sahukara, Krishna Bettouche, Zineddine Fischer, Andreas |
| author_facet | Sahukara, Krishna Bettouche, Zineddine Fischer, Andreas |
| contents | Document pages captured by smartphones or scanners often contain tables, yet manual extraction is slow and error-prone. We introduce an automated LaTeX-based pipeline that synthesizes realistic two-column pages with visually diverse table layouts and aligned ground-truth masks. The generated corpus augments the real-world Marmot benchmark and enables a systematic resolution study of TableNet. Training TableNet on our synthetic data achieves a pixel-wise XOR error of 4.04% on our synthetic test set with a 256x256 input resolution, and 4.33% with 1024x1024. The best performance on the Marmot benchmark is 9.18% (at 256x256), while cutting manual annotation effort through automation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_14583 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Synthetic Data Augmentation for Table Detection: Re-evaluating TableNet's Performance with Automatically Generated Document Images Sahukara, Krishna Bettouche, Zineddine Fischer, Andreas Computer Vision and Pattern Recognition Artificial Intelligence Document pages captured by smartphones or scanners often contain tables, yet manual extraction is slow and error-prone. We introduce an automated LaTeX-based pipeline that synthesizes realistic two-column pages with visually diverse table layouts and aligned ground-truth masks. The generated corpus augments the real-world Marmot benchmark and enables a systematic resolution study of TableNet. Training TableNet on our synthetic data achieves a pixel-wise XOR error of 4.04% on our synthetic test set with a 256x256 input resolution, and 4.33% with 1024x1024. The best performance on the Marmot benchmark is 9.18% (at 256x256), while cutting manual annotation effort through automation. |
| title | Synthetic Data Augmentation for Table Detection: Re-evaluating TableNet's Performance with Automatically Generated Document Images |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2506.14583 |