Parametric Synthesis of Text on Stylized Backgrounds using PGGANs
This work addresses the need for synthetic data generation in domain-specific applications like license plate recognition, but it is incremental as it builds on existing GAN and retrieval techniques.
The paper tackles the problem of generating high-resolution, photo-realistic images of text on stylized backgrounds by combining text-to-image retrieval with PGGANs, achieving conditional generation for automotive license plates and evaluating fidelity with license plate recognition systems.
We describe a novel method of generating high-resolution real-world images of text where the style and textual content of the images are described parametrically. Our method combines text to image retrieval techniques with progressive growing of Generative Adversarial Networks (PGGANs) to achieve conditional generation of photo-realistic images that reflect specific styles, as well as artifacts seen in real-world images. We demonstrate our method in the context of automotive license plates. We assess the impact of varying the number of training images of each style on the fidelity of the generated style, and demonstrate the quality of the generated images using license plate recognition systems.