Generative Models

Learning generative models that can explain complex data distribution is a long-standing problem in machine learning research. At SE(3), we are particularly interested in image generation, which is extremely challenging due to the high dimensionality of data. Generative models of images are not only important for unsupervised feature learning, but also enable a wide range of commercial applications such as image editing. With recent advances in Generative Adversarial Networks (GANs), it becomes possible to generate realistic images in constrained domains. Our research aims at improving and better understanding GANs, as well as exploring alternative methods for image generation.

Works

  • Arbitrary Style Transfer in Real-time with Adaptive Instance Normalization Gatys et al. recently introduced a neural algorithm that renders a content image in the style of another image, achieving so-called style transfer. However, their framework requires a slow iterative optimization process, which limits its practical application. Fast approximations with feed-forward neural networks have been proposed to speed up neural style transfer. Unfortunately, the speed ...
  • Precise Recovery of Latent Vectors from Generative Adversarial Networks Generative adversarial networks (GANs) transform latent vectors into visually plausible images. It is generally thought that the original GAN formulation gives no out-of-the-box method to reverse the mapping, projecting images back into latent space. We introduce a simple, gradient-based technique called stochastic clipping. In experiments, for images generated by the GAN, we precisely recover their ...
  • Stacked Generative Adversarial Networks In this paper, we propose a novel generative model named Stacked Generative Adversarial Networks (SGAN), which is trained to invert the hierarchical representations of a bottom-up discriminative network. Our model consists of a top-down stack of GANs, each learned to generate lower-level representations conditioned on higher-level representations. A representation discriminator is introduced at each feature ...

People