latexocr for math expression understanding

Posted Feb 10, 2025 Updated Feb 10, 2025

By 信雄湊

7 min read

pix2tex - LaTeX OCR

The goal of this project is to create a learning based system that takes an image of a math formula and returns corresponding LaTeX code.

Using the model

To run the model you need Python 3.7+

If you don’t have PyTorch installed. Follow their instructions here.

Install the package pix2tex:

pip install "pix2tex[gui]"

Model checkpoints will be downloaded automatically.

There are three ways to get a prediction from an image.

You can use the command line tool by calling pix2tex. Here you can parse already existing images from the disk and images in your clipboard.
Thanks to @katie-lim, you can use a nice user interface as a quick way to get the model prediction. Just call the GUI with latexocr. From here you can take a screenshot and the predicted latex code is rendered using MathJax and copied to your clipboard. Under linux, it is possible to use the GUI with gnome-screenshot (which comes with multiple monitor support). For other Wayland compositers, grim and slurp will be used for wlroots-based Wayland compositers and spectacle for KDE Plasma. Note that gnome-screenshot is not compatible with wlroots or Qt based compositers. Since gnome-screenshot will be preferred when available, you may have to set the environment variable SCREENSHOT_TOOL to grim or spectacle in these cases (other available values are gnome-screenshot and pil). If the model is unsure about the what’s in the image it might output a different prediction every time you click “Retry”. With the temperature parameter you can control this behavior (low temperature will produce the same result).
You can use an API. This has additional dependencies. Install via pip install -U "pix2tex[api]" and run
1 python -m pix2tex.api.run
to start a Streamlit demo that connects to the API at port 8502. There is also a docker image available for the API: https://hub.docker.com/r/lukasblecher/pix2tex
1 2 docker pull lukasblecher/pix2tex:api docker run --rm -p 8502:8502 lukasblecher/pix2tex:api
To also run the streamlit demo run
1 docker run --rm -it -p 8501:8501 --entrypoint python lukasblecher/pix2tex:api pix2tex/api/run.py
and navigate to http://localhost:8501/

Use from within Python

  
from PIL import Image
from pix2tex.cli import LatexOCR

img = Image.open('path/to/image.png')
model = LatexOCR()
print(model(img))

The model works best with images of smaller resolution. That’s why I added a preprocessing step where another neural network predicts the optimal resolution of the input image. This model will automatically resize the custom image to best resemble the training data and thus increase performance of images found in the wild. Still it’s not perfect and might not be able to handle huge images optimally, so don’t zoom in all the way before taking a picture.

Always double check the result carefully. You can try to redo the prediction with an other resolution if the answer was wrong.

Want to use the package?

I’m trying to compile a documentation right now.

Visit here: https://pix2tex.readthedocs.io/

Training the model

Install a couple of dependencies pip install "pix2tex[train]".

First we need to combine the images with their ground truth labels. I wrote a dataset class (which needs further improving) that saves the relative paths to the images with the LaTeX code they were rendered with. To generate the dataset pickle file run

python -m pix2tex.dataset.dataset --equations path_to_textfile --images path_to_images --out dataset.pkl

To use your own tokenizer pass it via --tokenizer (See below).

You can find my generated training data on the Google Drive as well (formulae.zip - images, math.txt - labels). Repeat the step for the validation and test data. All use the same label text file.

Edit the data (and valdata) entry in the config file to the newly generated .pkl file. Change other hyperparameters if you want to. See pix2tex/model/settings/config.yaml for a template.
Now for the actual training run

python -m pix2tex.train --config path_to_config_file

If you want to use your own data you might be interested in creating your own tokenizer with

python -m pix2tex.dataset.dataset --equations path_to_textfile --vocab-size 8000 --out tokenizer.json

Don’t forget to update the path to the tokenizer in the config file and set num_tokens to your vocabulary size.

Model

The model consist of a ViT [1] encoder with a ResNet backbone and a Transformer [2] decoder.

Performance

BLEU score	normed edit distance	token accuracy
0.88	0.10	0.60

Data

We need paired data for the network to learn. Luckily there is a lot of LaTeX code on the internet, e.g. wikipedia, arXiv. We also use the formulae from the im2latex-100k [3] dataset. All of it can be found here

Dataset Requirements

In order to render the math in many different fonts we use XeLaTeX, generate a PDF and finally convert it to a PNG. For the last step we need to use some third party tools:

XeLaTeX
ImageMagick with Ghostscript. (for converting pdf to png)
Node.js to run KaTeX (for normalizing Latex code)
Python 3.7+ & dependencies (specified in setup.py)

Fonts

Latin Modern Math, GFSNeohellenicMath.otf, Asana Math, XITS Math, Cambria Math

TODO

add more evaluation metrics
create a GUI
add beam search
support handwritten formulae (kinda done, see training colab notebook)
reduce model size (distillation)
find optimal hyperparameters
tweak model structure
fix data scraping and scrape more data
trace the model (#2)

Contribution

Contributions of any kind are welcome.

Acknowledgment

Code taken and modified from lucidrains, rwightman, im2markup, arxiv_leaks, pkra: Mathjax, harupy: snipping tool

References

[1] An Image is Worth 16x16 Words

[2] Attention Is All You Need

[3] Image-to-Markup Generation with Coarse-to-Fine Attention

Installation issue

To resolve the installation issue, you need to reinstall Pillow with XCB support enabled. Follow these steps:

Install Required Dependencies: Ensure that the necessary development libraries are installed on your system. For Ubuntu-based systems, you can install them using the following command:

  
sudo apt-get install libtiff5-dev libjpeg8-dev libopenjp2-7-dev zlib1g-dev \
libfreetype6-dev liblcms2-dev libwebp-dev tcl8.6-dev tk8.6-dev python3-tk \
libharfbuzz-dev libfribidi-dev libxcb1-dev

These packages provide the necessary headers and libraries for compiling Pillow with support for various image formats and functionalities. citeturn0search2

Reinstall Pillow: After installing the dependencies, reinstall Pillow using pip with the --no-binary option to force a source build:
1 python3 -m pip install --upgrade Pillow --no-binary :all:
This command compiles Pillow from source, ensuring that it includes support for XCB and other necessary libraries. citeturn0search2

Alternative Solution: Use Precompiled Binaries

If you prefer not to compile from source, you can use precompiled Pillow binaries that include XCB support. However, availability may vary depending on your operating system and Python version.

Additional Considerations

Virtual Environments: If you’re using a virtual environment, ensure that it’s activated when installing the dependencies and reinstalling Pillow. This ensures that the correct environment is used.
System Architecture: Ensure that the libraries and binaries you install are compatible with your system’s architecture (e.g., x86_64).
Python Version: Verify that the Python version in your virtual environment matches the one used when installing the system packages.

By following these steps, the OSError related to missing XCB support in Pillow will be solved.

To resolve this issue, you need to reinstall Pillow with XCB support enabled. Follow these steps: