Fast-converging Conditional Generative Adversarial Networks for Image Synthesis
This work addresses convergence speed issues in conditional GANs for image synthesis, offering an incremental improvement over existing methods like AC-GAN.
The paper tackled the problem of slow convergence in conditional GANs for image synthesis by proposing FC-GAN, which uses an advanced auxiliary classifier to avoid mixing generated and real data, resulting in faster convergence while maintaining competitive image quality.
Building on top of the success of generative adversarial networks (GANs), conditional GANs attempt to better direct the data generation process by conditioning with certain additional information. Inspired by the most recent AC-GAN, in this paper we propose a fast-converging conditional GAN (FC-GAN). In addition to the real/fake classifier used in vanilla GANs, our discriminator has an advanced auxiliary classifier which distinguishes each real class from an extra `fake' class. The `fake' class avoids mixing generated data with real data, which can potentially confuse the classification of real data as AC-GAN does, and makes the advanced auxiliary classifier behave as another real/fake classifier. As a result, FC-GAN can accelerate the process of differentiation of all classes, thus boost the convergence speed. Experimental results on image synthesis demonstrate our model is competitive in the quality of images generated while achieving a faster convergence rate.