DeLiGAN : Generative Adversarial Networks for Diverse and Limited Data
This addresses the challenge of data scarcity in generative modeling for image synthesis, though it is an incremental improvement over existing GAN methods.
The authors tackled the problem of generating diverse images with limited training data by proposing DeLiGAN, a GAN-based architecture that reparameterizes the latent space as a mixture model, resulting in effective diversity in generated samples for handwritten digits, objects, and sketches using limited data.
A class of recent approaches for generating images, called Generative Adversarial Networks (GAN), have been used to generate impressively realistic images of objects, bedrooms, handwritten digits and a variety of other image modalities. However, typical GAN-based approaches require large amounts of training data to capture the diversity across the image modality. In this paper, we propose DeLiGAN -- a novel GAN-based architecture for diverse and limited training data scenarios. In our approach, we reparameterize the latent generative space as a mixture model and learn the mixture model's parameters along with those of GAN. This seemingly simple modification to the GAN framework is surprisingly effective and results in models which enable diversity in generated samples although trained with limited data. In our work, we show that DeLiGAN can generate images of handwritten digits, objects and hand-drawn sketches, all using limited amounts of data. To quantitatively characterize intra-class diversity of generated samples, we also introduce a modified version of "inception-score", a measure which has been found to correlate well with human assessment of generated samples.