Hugging Face's logo

Datasets:
openpecha
/
Tibetan-header-footer

Modalities:
Image
Text
Dataset card Data Studio Files
xet
 Community

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.

YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

Tibetan Header/Footer (YOLO detection dataset)

This dataset contains page images annotated for header/footer-related page layout elements.

Hosted on Hugging Face: openpecha/Tibetan-header-footer.

Contents

  • Task: object detection
  • Format: YOLO text labels (class x_center y_center width height, all normalized to [0,1])
  • Images: 313 total
    • Train: 251
    • Val: 62
  • Classes (nc=4):
    • 0: header
    • 1: footer
    • 2: footnote
    • 3: page_number

Folder structure 

images/
  train/   # .jpg
  val/     # .jpg
label/
  train/   # .txt (YOLO)
  val/     # .txt (YOLO)
data.yaml

How it was produced

Annotations were not created purely by hand from scratch.

  • Step 1 (model inference): bounding boxes were generated by running inference on the page images using:
    • DocLayout-YOLO
    • PP-DocLayout
  • Step 2 (manual correction): the model outputs were then manually corrected using the Ultralytics annotation web app (Ultralytics Hub).

Notes

  • Each image X.jpg has a corresponding label file label/{train|val}/X.txt.
  • Images may contain multiple boxes across classes.

License

MIT

license: mit

Downloads last month
6