Loading...

Disentangled Representation Learning for Automated Clothe Image Synthesis on the Body

Johary, Iman | 2022

199 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 54832 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Hashemi, Matin
  7. Abstract:
  8. There have been many works on generative networks and image generation in the past few years, but the problem with this work is that there is no control over the generated images. The goal of disentangled image synthesis is to generate new images with specific detail and have control over the generated images. Image-based virtual try-on aims to synthesize the customer image with an in-shop clothes image to acquire seamless and natural try-on results, which have attracted increasing attention. The main procedures of image-based virtual try-on usually consist of clothes image generation and try-on image synthesis. In contrast, prior arts cannot guarantee satisfying clothes results when facing significant geometric changes and complex clothes patterns, which further deteriorates the afterward try-on results. To address this issue, we propose a virtual try-on network based on segmentation-based shape matching. Specifically, the clothes image generation progressively learns the warped clothes and refined clothes in an end-to-end manner. We introduce a segmentation-based constraint in Thin-Plate Spline (TPS) warping to change the problem from a complex problem to five smaller warping problems resistant to significant geometric changes. To quantify image synthesis quality, we introduce a virtual try-on metric called cloth reconstruction metric for the first time
  9. Keywords:
  10. Representation Learning ; Semantic Segmentation ; Thin Plate Spline ; Generative Adversarial Networks ; Image Synthesis ; Virtual Try-On

 Digital Object List

 Bookmark

No TOC