Handwritten Signature Spoofing With Conditional Generative Adversarial Nets

Generative adversarial networks have been a highly focused research topic in computer vision, especially in image synthesis and image-to-image translation. There are a lot of variations in generative nets, and different GANs are suitable for different applications. In this chapter, the authors investigated conditional generative adversarial networks to generate fake images, such as handwritten signatures. The authors demonstrated an implementation of conditional generative adversarial networks, which can generate fake handwritten signatures according to a condition vector tailored by humans.