File size: 2,756 Bytes
2623ea7
 
dfb1e24
23006e2
2623ea7
 
23006e2
2623ea7
 
 
 
 
 
 
 
 
 
23006e2
2623ea7
 
23006e2
 
 
2623ea7
 
 
 
23006e2
2623ea7
23006e2
 
 
 
2623ea7
23006e2
 
 
 
 
 
 
 
2623ea7
 
 
23006e2
2623ea7
 
 
 
 
 
 
 
 
12c159e
2623ea7
 
 
 
 
 
23006e2
 
 
2623ea7
 
 
 
 
 
23006e2
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
---
library_name: transformers
pipeline_tag: image-to-text
license: afl-3.0
---

# Model Card for TrOCR_Math_handwritten

<!-- Provide a quick summary of what the model is/does. -->



## Model Details


<!-- Provide a longer summary of what this model is. -->

TrOCR model fine-tuned on a part of the [mathwriting](https://github.com/google-research/google-research/tree/master/mathwriting) dataset converted from InkML files into images. It was introduced in the paper [TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models](https://arxiv.org/abs/2109.10282) by Li et al. and first released in [this repository](https://github.com/microsoft/unilm/tree/master/trocr).

- **Developed by:** [More Information Needed]
- **Model type:** Transformer OCR
- **License:** afl-3.0
- **Finetuned from model [optional]:** [TrOCR_large_stage1](https://huggingface.co/microsoft/trocr-large-stage1)


## Uses

Here is how to use this model in PyTorch:

```python
from transformers import TrOCRProcessor, VisionEncoderDecoderModel
from PIL import Image
import requests

url = "path/to/image"
image = Image.open(requests.get(url, stream=True).raw).convert("RGB")
processor = TrOCRProcessor.from_pretrained('fhswf/TrOCR_Math_handwritten')
model = VisionEncoderDecoderModel.from_pretrained('fhswf/TrOCR_Math_handwritten')
pixel_values = processor(images=image, return_tensors="pt").pixel_values
generated_ids = model.generate(pixel_values)
generated_text = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]
```

## Bias, Risks, and Limitations

You can use the raw model for optical character recognition (OCR) on images containing one mathematical formula.



## Training Details

### Training Data

<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->

This model was finetuned on a part of the [mathwriting](https://github.com/google-research/google-research/tree/master/mathwriting) dataset converted from InkML files into images.



## Evaluation

<!-- This section describes the evaluation protocols and provides the results. -->
Percentage of correct recognition: 77.8% <br>
Percentage of correct recognition with one error: 85.7% <br>
Percentage of correct recognition with two error: 89.9%




**BibTeX:**

```bibtex
@misc{li2021trocr,
      title={TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models}, 
      author={Minghao Li and Tengchao Lv and Lei Cui and Yijuan Lu and Dinei Florencio and Cha Zhang and Zhoujun Li and Furu Wei},
      year={2021},
      eprint={2109.10282},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}
```