NORMA eResearch @NCI Library

Fashion Outfit Design Image Synthesis Using Comparative Study of Generative Adversarial Networks

Jain, Karan (2020) Fashion Outfit Design Image Synthesis Using Comparative Study of Generative Adversarial Networks. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
PDF (Master of Science)
Download (2MB) | Preview
[thumbnail of Configuration manual]
PDF (Configuration manual)
Download (272kB) | Preview


Despite of the capabilities and effects of Artificial Intelligence (AI) it is been observed that there are certain industries which are yet to utilize the immense ability of AI. Fashion industry is slowly making use of machine learning which was used to be a less touched domain of analytics. AI and data analytics both together can bring innovations in the fashion world. A part of this fashion industry is very difficult and that is the fashion designing where new ideas and creations need to be brought in through a line of work. The research is helpful for the fashion designers as it decreases their work pressure using AI to develop fashion images. The Generative Adversarial Network (GAN) is a model to generate samples of images, videos, texts etc. which are more naturalistic. This research focuses on comparative examination using two types of GAN viz- Deep Convolutional Neural Network based GAN (DCGAN) and the Capsule Network based GAN (CapsGAN) to develop new and distinctive images of fashion outfits, jewelries, shoes etc. of both men and women. Both Qualitative and Quantitative methods been used to evaluate the generated images. This resulted in better images generated by CapsGAN than DCGAN. DCGAN showed higher Discriminator loss and in contrast to this CapsGAN showed rise in Generator loss. Altogether it can be stated that CapsGAN performance is much better than DCGAN.
Index terms— Generative Adversarial Networks, Deep Convolutional GAN, Capsule Network GAN, Fashion Industry, Deep Learning

Item Type: Thesis (Masters)
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
Q Science > QA Mathematics > Computer software
T Technology > T Technology (General) > Information Technology > Computer software
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Dan English
Date Deposited: 20 Jan 2021 15:36
Last Modified: 20 Jan 2021 15:36

Actions (login required)

View Item View Item