GoBo - A Handwriting Recognition dataset for Personalization

This dataset comprises the images for the personalization described in the paper Personalizing Handwriting Recognition Systems with Limited User-Specific Samples.

Dataset Statistics (v.1.0)

* Handwritten word-level images
* English
* 40 Participants
* 5 sets from different sources for personalization 
* 2 sets from 2 domains (same domains as 2 personalization sets) for testing
* 926 words/writer, 37k words in total

More details can be found on the Github Repository:
Github GoBo

Model
gobo_Baselinemodel.hdf5