Personal tools
You are here: Home Dataset CCC

Cursive Character Challenge

This is the home page of Cursive Character Challenge (C-Cube), the new benchmark for machine learning and pattern recognition algorithms. The database contains 57293 cursive characters manually extracted from cursive words, including both upper and lower case versions of each letter.
The database is split into:

- training set (38160 characters)
- test set (19133 characters)

To download database, click here. The database provides experimental protocols in order to compare rigorously the results obtained by different researchers.

The quality of this site depends on you!! If you have any suggestion or comment do not hesitate to contact us:

- Francesco Camastra (
- Marco Spinetti (
- Alessandro Vinciarelli