Datasets

This page lists the ICDAR 2003 datasets available for download on this site: Robust Reading , Robust Word Recognition , Robust OCR , Text Locating and Cursive Script .

Please note that the Page Segmentation and Table Segmentation competitions have their own separate datasets and procedures.

Robust Reading Datasets

These datasets were collected and tagged by the ICDAR 2003 Robust Reading Dataset Collection Team ( photo .  Clockwise from left: Shirley Wong, Simon Lucas, Alex Panaretos, Luis Sosa Velazquez, Robert Young, Anthony Tang.)

The datasets are organized into Sample , Trial and Competition datasets.

Sample datasets are provided to give you a quick impression of the data, and also to allow function testing of your software.  That is, you can run tests on the sample data to check that your software works with the data, but the results won't mean much.

Trial datasets serve two purposes.  Use them to get results for your ICDAR 2003 papers.  For this purpose, they are partitioned into two sets: TrialTrain and TrialTest.  Use TrialTrain to train or tune your algorithms, then quote results on TrialTest.  For the competitions, you should train/tune your system on the entire Trial set. 

Competition datasets will be used to measure the performance of your algorithms for the competitions.  These will be kept private until the ICDAR 2003 conference, when they will be made public.

Robust Reading and Text Locating

• Sample (20 images, 7.3mb)

• TrialTrain (258 images, 43.3mb)

• TrialTest (251 images, 69.6mb)

Robust Word Recognition

• Sample (171 words, 2.7mb)

• TrialTrain (1157 words, 17.5mb)

• TrialTest (1111 words , 19.6mb)

Robust OCR

• Sample (854 chars, 3.0mb)

• TrialTrain (6185 chars, 22.3mb)

• TrialTest (5430 chars , 23.9mb)

Cursive Script Recognition

Off-line cursive script datasets are available here .