Skip to content

DCGM/semANT-TextBite

Repository files navigation

semANT-TextBite

Tool for extracting logical chunks from complex document pages.

Install from PyPi:

pip install textbite

Run by simply providing a folder of images alongside a folder of corresponding PAGE XMLs (such as obtained from pero-ocr):

textbite \
    --xml xmls/ \
    --img imgs/ \
    --yolo yolo.pt \
    --gnn gnn.pth \
    --normalizer norm.pkl \
    --save out/ \
    [--logging-level INFO] \
    [--json] \
    [--alto altos/]

By default, TextBite downloads neccessary models from the internet. In case you are working in an offline environment, you can download it yourself and provide path as an argument.

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages