Open Source Handwriting Recognition

An alternative approach to handwriting recognition without ML
Anish Sheela

Anish Sheela

Handwriting recognition is traditionally done using Machine Learning and often require huge set of training data and is often proprietary. Anish will be demonstrating an algorithmic approach using procustes analysis which can recognize the written handwriting input using curve matching against a stored shape. It currently works for Malayalam and Tamil as a proof of concept and requires a lot of improvement. It can be extended to all Indian languages and can be used as a typing tool for Mobile, which values your privacy.
Agenda Code of Conduct Contact Us

Organized and Hosted by the folks at

Zerodha Logo


Frappe Logo