Text and Style Conditioned GAN for Generation of Offline Handwriting Lines
This addresses the need for realistic handwriting generation in applications like document forgery detection or personalized fonts, though it is incremental over prior single-word methods.
The paper tackles the problem of generating entire lines of offline handwriting images conditioned on text and style, achieving results that human evaluators found indistinguishable from human-written handwriting.
This paper presents a GAN for generating images of handwritten lines conditioned on arbitrary text and latent style vectors. Unlike prior work, which produce stroke points or single-word images, this model generates entire lines of offline handwriting. The model produces variable-sized images by using style vectors to determine character widths. A generator network is trained with GAN and autoencoder techniques to learn style, and uses a pre-trained handwriting recognition network to induce legibility. A study using human evaluators demonstrates that the model produces images that appear to be written by a human. After training, the encoder network can extract a style vector from an image, allowing images in a similar style to be generated, but with arbitrary text.