Skip to content

Latest commit

 

History

History
17 lines (10 loc) · 1.24 KB

README.md

File metadata and controls

17 lines (10 loc) · 1.24 KB
Raw

Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks

This project is part of course CS F425 Deep Learning at BITS Goa.

In this project, we learned that Cycle Consistency Adversarial Networks tries to solve image-image translation problems without paired data. The objective function it uses is the Adversarial losses and Cycle consistency losses. Image-Image translation is used in Object Transfiguration, Photo Enhancement, Style Transformation, Season Transformation and Image Segmentation. We made the model and tested it on a new city dataset and performed image segmentation.

Addtional details about the project can be found in project report.

Team Members

  • Anurag Nagpal [2018B1A70939G]
  • Sahdev Nuwal [2018B3A70900G]
  • Varun Bankar [2018B2A70295G]

You can find Colab notebook related to this project here.

This project is based on the Paper titled “Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks” by Jun-Yan Zhu et al 2017.