Sorry, you need to enable JavaScript to visit this website.

DATA INCUBATION — SYNTHESIZING MISSING DATA FOR HANDWRITING RECOGNITION

Citation Author(s):
Jen-Hao Rick Chang, Martin Bresler, Youssouf Chherawala, Adrien Delaye, Thomas Deselaers, Ryan Dixon, Oncel Tuzel
Submitted by:
Jen-Hao Rick Chang
Last updated:
4 May 2022 - 5:35pm
Document Type:
Presentation Slides
Event:
Presenters:
Jen-Hao Rick Chang
Paper Code:
3765
Categories:
 

In this paper, we demonstrate how a generative model can be used to build a better recognizer through the control of content and style. We are building an online handwriting recognizer from a modest amount of training samples. By training our controllable handwriting synthesizer on the same data, we can synthesize handwriting with previously underrepresented content (e.g., URLs and email addresses) and style (e.g., cursive and slanted). Moreover, we propose a framework to analyze a recognizer that is trained with a mixture of real and synthetic training data. We use the framework to optimize data synthesis and demonstrate significant improvement on handwriting recognition over a model trained on real data only. Overall, we achieve a 66% reduction in Character Error Rate.

up
1 user has voted: Jen-Hao Rick Chang