README.md

---
language:
- en
- hi
license: apache-2.0
task_categories:
- image-to-text
tags:
- table-structure-recognition
- tsr
- document-understanding
- ocr
- benchmark
- indian-textbooks
- html
- sample
size_categories:
- n<1K
---

# DrishtiTable: A Benchmark for Table Structure Recognition in Indian Academic Textbooks

> **Try the model now!** Upload any table image and get HTML structure back:
> **[DrishtiTable Live Demo](https://huggingface.co/spaces/nalanda-data/DrishtiTable)**

> **This is a sample release** containing 20 representative tables (10 train / 5 val / 5 test). The full dataset contains 1,421 tables. Contact the authors for full dataset access.

DrishtiTable is a curated dataset of table images with high-quality HTML structure annotations from Indian academic textbooks published by S. Chand Publications. It serves as a benchmark for evaluating Table Structure Recognition (TSR) models on domain-specific educational content.

## Benchmark Results (Full Dataset)

| Model | Method | TEDS | S-TEDS |
|---|---|---|---|
| Qwen2.5-VL-7B | Zero-shot | 58.8% | 74.0% |
| o4-mini (OpenAI) | Zero-shot | 61.4% | 70.0% |
| GPT-4.1 (OpenAI) | Zero-shot | 68.0% | 80.8% |
| GPT-4o (OpenAI) | Zero-shot | 71.1% | 84.3% |
| **DrishtiTable-Qwen2.5-VL-7B** | **SFT** | **83.2%** | **89.7%** |

## Sample Dataset Statistics

| Split | Samples (this release) | Samples (full dataset) |
|---|---|---|
| Train | 10 | 1,141 |
| Validation | 5 | 145 |
| Test | 5 | 135 |
| **Total** | **20** | **1,421** |

## Full Dataset Statistics

The complete DrishtiTable dataset spans 9 Indian academic textbooks across 6 subjects:

| Subject | Tables |
|---|---|
| Financial Accounting | 780 |
| Business Statistics | 261 |
| Quantitative Techniques | 183 |
| Operation Research | 108 |
| Steam Tables (Engineering) | 52 |
| Ethics | 37 |

### Complexity Features (Full Dataset)

| Feature | Count | Percentage |
|---|---|---|
| Merged cells (colspan/rowspan) | 200 | 14.1% |
| Multi-level hierarchy | 219 | 15.4% |
| Empty cells | 340 | 23.9% |
| Bold text | 890 | 62.6% |

## Data Format

Each sample consists of:
- **Image**: Cropped table region as PNG (`images/`)
- **Annotation**: HTML structure with semantic tags (`annotations/`)
- **Metadata**: Table properties as JSON (`metadata/`)

### HTML Annotation Example

```html
<table>
  <thead>
    <tr>
      <th>Year</th>
      <th colspan="2">Revenue</th>
    </tr>
  </thead>
  <tbody>
    <tr>
      <td>2024</td>
      <td>1,234</td>
      <td>5,678</td>
    </tr>
  </tbody>
</table>
```

## File Structure

```
DrishtiTable/
  images/           # 20 PNG table images (sample)
  annotations/      # 20 HTML files
  metadata/         # 20 JSON metadata files
  train.csv         # 10 training samples
  val.csv           # 5 validation samples
  test.csv          # 5 test samples
```

## Fine-Tuned Model and Demo

| Resource | Link |
|---|---|
| **Live Demo** | [DrishtiTable Space](https://huggingface.co/spaces/nalanda-data/DrishtiTable) |
| **Fine-tuned Model** | [Nalandadata/DrishtiTable-Qwen2.5-VL-7B](https://huggingface.co/Nalandadata/DrishtiTable-Qwen2.5-VL-7B) |
| **Base Model** | [Qwen/Qwen2.5-VL-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-VL-7B-Instruct) |

## Citation

```bibtex
@article{drishtitable2026,
  title={Domain-Specific Fine-Tuning for Table Structure Recognition: A 7B Open Model Outperforms GPT-4o with 1,141 Training Samples},
  author={Nalanda Data},
  year={2026}
}
```

## Commercial Use & Licensing

This dataset is released under **Apache 2.0**, which permits commercial use under the terms of that license.

The public release on Hugging Face is a sample of **20 tables**. The full DrishtiTable corpus contains **1,421 expert-annotated tables** across 9 Indian academic textbooks, spanning Financial Accounting, Business Statistics, Quantitative Techniques, Operations Research, Steam Tables, and Ethics.

For commercial users who need:

- **Full dataset access** — 1,421 tables across 6 subjects
- **Custom slices** — specific subjects, table complexity profiles, multi-page layouts
- **Custom data work** — additional textbook collection, annotation, evaluation pipelines
- **Production deployment support** for fine-tuned TSR models

### Contact

**For commercial licensing, full dataset access, custom data work, or partnerships:**  
📧 [info@nalandadata.ai](mailto:info@nalandadata.ai)

**For technical questions, integration help, or fine-tuning support:**  
📧 [tech@nalandadata.ai](mailto:tech@nalandadata.ai)

🌐 [nalandadata.ai](https://nalandadata.ai)

## License

Apache 2.0