Handwriting Profiling using Generative Adversarial Networks

TL;DR

This paper introduces a modified DCGAN architecture combined with reinforcement learning to learn and mimic handwriting, with applications in forgery detection, signature verification, and digitization, demonstrated on MNIST data.

Contribution

It presents a novel GAN-based approach with reinforcement learning for handwriting profiling, enhancing learning speed and accuracy over existing methods.

Findings

Good performance on MNIST dataset

Potential applications in forgery detection and digitization

Enhanced learning speed with reinforcement learning

Abstract

Handwriting is a skill learned by humans from a very early age. The ability to develop one's own unique handwriting as well as mimic another person's handwriting is a task learned by the brain with practice. This paper deals with this very problem where an intelligent system tries to learn the handwriting of an entity using Generative Adversarial Networks (GANs). We propose a modified architecture of DCGAN (Radford, Metz, and Chintala 2015) to achieve this. We also discuss about applying reinforcement learning techniques to achieve faster learning. Our algorithm hopes to give new insights in this area and its uses include identification of forged documents, signature verification, computer generated art, digitization of documents among others. Our early implementation of the algorithm illustrates a good performance with MNIST datasets.