Products

The research material listed below is available free to the general population. It was created in our lab or in collaborations our lab has had with other groups. 

Code

Character Spotting & AutoTag – Explanation and code to create YOLO training sets for handwritten data using AutoTag. Based on work by Nishatul Majid during his PhD study.

Data

Boise State Bangla Handwriting Dataset – characters and essays in Bangla text used in work by PhD student Nishatul Majid 

Dataset for Normalization of Unconstrained Handwritten Words in Terms of Slope and Slant Correction  –  Multilingual datasets (Bangla and Devanagari) comprising both real and synthetic handwritten word images along with ground truth information related to the writing's slope and slant.

Material from Papers & Books

Reconnaissance des Formes, Théorie et Pratique sous matlab – datasets and material for L. Likforman-Sulem and E. H. Barney Smith, Reconnaissance des Formes, Théorie et Pratique sous matlab, Technosup Collection, Ellipses Édition Marketing, 2013.