GAN-based Algorithm for Efficient Image Inpainting

TL;DR

This paper presents a GAN-based image inpainting method combining autoencoders and context encoders to restore masked faces, demonstrating promising results with potential for future improvements.

Contribution

The paper introduces a novel hybrid model combining autoencoders and GANs for face inpainting, trained on influencer images, with discussions on its limitations and future directions.

Findings

Effective face inpainting with GAN-autoencoder hybrid

Model trained on 50,000 influencer images

Identifies areas for model improvement and future research

Abstract

Global pandemic due to the spread of COVID-19 has post challenges in a new dimension on facial recognition, where people start to wear masks. Under such condition, the authors consider utilizing machine learning in image inpainting to tackle the problem, by complete the possible face that is originally covered in mask. In particular, autoencoder has great potential on retaining important, general features of the image as well as the generative power of the generative adversarial network (GAN). The authors implement a combination of the two models, context encoders and explain how it combines the power of the two models and train the model with 50,000 images of influencers faces and yields a solid result that still contains space for improvements. Furthermore, the authors discuss some shortcomings with the model, their possible improvements, as well as some area of study for future investigation for applicative perspective, as well as directions to further enhance and refine the model.