Multiple Generative Adversarial Networks Analysis for Predicting Photographers' Retouching

TL;DR

This paper investigates the use of multiple GAN models to automatically retouch photos, aiming to replicate professional editing and identify the most effective GAN architecture for this task.

Contribution

It explores various GAN architectures for photo retouching and evaluates their effectiveness in mimicking expert-level editing.

Findings

Certain GAN models outperform others in retouching quality.

Deep learning can approximate professional photo retouching.

The study identifies the most suitable GAN architecture for this application.

Abstract

Anyone can take a photo, but not everybody has the ability to retouch their pictures and obtain result close to professional. Since it is not possible to ask experts to retouch thousands of pictures, we thought about teaching a piece of software how to reproduce the work of those said experts. This study aims to explore the possibility to use deep learning methods and more specifically, generative adversarial networks (GANs), to mimic artists' retouching and find which one of the studied models provides the best results.