Structured GANs
This work addresses the need for controllable image generation in computer vision, particularly for face synthesis, but it is incremental as it builds on existing GAN frameworks.
The paper tackles the problem of controlling symmetry in generated images by introducing Structured GANs, which modify the generator architecture while keeping training and loss unchanged, resulting in the ability to generate novel faces with varying symmetry and achieve unsupervised face rotation.
We present Generative Adversarial Networks (GANs), in which the symmetric property of the generated images is controlled. This is obtained through the generator network's architecture, while the training procedure and the loss remain the same. The symmetric GANs are applied to face image synthesis in order to generate novel faces with a varying amount of symmetry. We also present an unsupervised face rotation capability, which is based on the novel notion of one-shot fine tuning.