Hugging Face's logo

Datasets:
cis-lmu
/
uniOCR.bench

Tasks:
Image-to-Text
Modalities:
Image
Text
Formats:
parquet
Languages:
English
Russian
Chinese
Size:
10K - 100K
Tags:
text
image
ocr
Libraries:
Datasets
Dask
Polars
License:
Dataset card Data Studio Files
xet
Community
1

You need to agree to share your contact information to access this dataset

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Log in or Sign Up to review the conditions and access this dataset content.

uniOCR.bench

uniOCR.bench is a dataset of X images covering Y writing systems (X languages), designed to evaluate the fundamental OCR capabilities required to support diverse writing systems and languages.

Quick links:

License

  • The uniOCR.bench metadata is licensed under CC0-1.0. The texts used to generate the images come from the GlotLID repository. Each text is accompanied by a source tag here and is licensed under its own terms. You can find the list of sources (match the key) and their licenses here.
  • The licenses of the all source texts permit their use for evaluation. This dataset is intended only for evaluation and must not be used for training, for either commercial or non-commercial purposes.
  • The image generation code is released under the Apache 2.0 license and may be used to create training data, provided that different seed texts from those used in our evaluation dataset are used.
  • For specific use cases, please contact us (amir@cis.lmu.de).
Downloads last month
151