Data Incubation -- Synthesizing Missing Data for Handwriting Recognition
This work addresses data scarcity for handwriting recognition systems, offering a practical solution for improving accuracy in applications like document digitization.
The paper tackles the problem of limited training data for online handwriting recognition by using a generative model to synthesize missing content and styles, achieving a 66% reduction in Character Error Rate.
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.