Unsupervised Image Generation with Infinite Generative Adversarial Networks
This addresses problems in unsupervised image generation for computer vision researchers, though it appears incremental as it builds on existing GAN frameworks.
The paper tackles issues like mode collapse and unstructured latent spaces in GANs for image generation by proposing MIC-GANs, an unsupervised non-parametric method that outperforms state-of-the-art methods in evaluations across datasets.
Image generation has been heavily investigated in computer vision, where one core research challenge is to generate images from arbitrarily complex distributions with little supervision. Generative Adversarial Networks (GANs) as an implicit approach have achieved great successes in this direction and therefore been employed widely. However, GANs are known to suffer from issues such as mode collapse, non-structured latent space, being unable to compute likelihoods, etc. In this paper, we propose a new unsupervised non-parametric method named mixture of infinite conditional GANs or MIC-GANs, to tackle several GAN issues together, aiming for image generation with parsimonious prior knowledge. Through comprehensive evaluations across different datasets, we show that MIC-GANs are effective in structuring the latent space and avoiding mode collapse, and outperform state-of-the-art methods. MICGANs are adaptive, versatile, and robust. They offer a promising solution to several well-known GAN issues. Code available: github.com/yinghdb/MICGANs.