ePrints@IIScePrints@IISc Home | About | Browse | Latest Additions | Advanced Search | Contact | Help

DeLiGAN : Generative Adversarial Networks for Diverse and Limited Data

Gurumurthy, Swaminathan and Sarvadevabhatla, Ravi Kiran and Babu, RV (2017) DeLiGAN : Generative Adversarial Networks for Diverse and Limited Data. In: 30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017), JUL 21-26, 2016, Honolulu, HI, pp. 4941-4949.

[img] PDF
Ieee_Con_Com_4941_2017.pdf - Published Version
Restricted to Registered users only

Download (1MB) | Request a copy
Official URL: http://dx.doi.org/10.1109/CVPR.2017.525

Abstract

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.

Item Type: Conference Poster
Series.: IEEE Conference on Computer Vision and Pattern Recognition
Publisher: 10.1109/CVPR.2017.525
Additional Information: 30th IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, JUL 21-26, 2016 Copy right for this article belongs to the IEEE, 345 E 47TH ST, NEW YORK, NY 10017 USA
Department/Centre: Others
Date Deposited: 20 Jan 2018 05:40
Last Modified: 25 Feb 2019 05:49
URI: http://eprints.iisc.ac.in/id/eprint/58846

Actions (login required)

View Item View Item