karimgamaleldin / adi-gan-s

Adi-GAN-s

Your faithful shoe designer

Adi-GAN-s is a project that uses the edges2shoes dataset to transform shoe sketches into detailed designs using Generative Adversarial Networks (GANs).

Introduction

Adi-GAN-s leverages the power of GANs to create realistic shoe designs from simple sketches. This project is based on the pix2pix framework and utilizes the edges2shoes dataset for training and evaluation.

Tech Stack

Results

Here are some example results generated by Adi-GAN-s:

Example 1

Example 2

Example 3

Example 4

Example 5

Example 6

Acknowledgements

• Image-to-Image Translation with Conditional Adversarial Networks
• Aladdin Persson