Handwritten Annotation Detection Dataset (AnnotationDB)

2018-05-31 (v. 1)

Contact author

Andreas Kölsch

TU Kaiserslautern

a_koelsch12@cs.uni-kl.de

-

You can cite this dataset as: Andreas Kölsch, Handwritten Annotation Detection Dataset (AnnotationDB) ,1,ID:AnnotationDB_1,URL:http://tc11.cvc.uab.es/datasets/AnnotationDB_1

Dataset Information

Keywords

Handwriting, Annotation, Segmentation, Historic, German, Documents

Description

The dataset contains 40 images for training and validation and 10 images for testing.

The document pages in the dataset are from multiple sources which are digitized using different devices. This increased variance makes the dataset especially challenging for segmentation task.

Technical Details

All images are labeled with their respective ground truths which are available in the PAGE format and as PNG files. The PNG files encode the classes in the Blue color channel and allow for ambiguous regions (cf. ICDAR2017 Competition on Layout Analysis for Challenging Medieval Manuscripts).

FileTypeSizeDownloadsDescription
test.zipdata(38 MB)10Test images
train.zipdata(111 MB)8Training images

Comments

No comments on this dataset yet.

Add your comment

In order to comment on a dataset you need to be logged on
Register Now!

Valoration

In order to rate this dataset you need to be logged on
Register Now!