JDSR-GAN: Constructing An Efficient Joint Learning Network for Masked Face Super-Resolution

TL;DR

This paper introduces JDSR-GAN, a joint learning network that effectively performs masked face super-resolution by combining denoising and super-resolution tasks, improving quality and identity preservation in low-resolution masked face images.

Contribution

The paper proposes a novel joint learning framework, JDSR-GAN, that simultaneously handles masked face denoising and super-resolution, outperforming separate task models.

Findings

JDSR-GAN achieves superior image quality over comparable methods.

The joint approach enhances identity preservation in reconstructed faces.

Incorporating attention and identity features improves feature learning.

Abstract

With the growing importance of preventing the COVID-19 virus, face images obtained in most video surveillance scenarios are low resolution with mask simultaneously. However, most of the previous face super-resolution solutions can not handle both tasks in one model. In this work, we treat the mask occlusion as image noise and construct a joint and collaborative learning network, called JDSR-GAN, for the masked face super-resolution task. Given a low-quality face image with the mask as input, the role of the generator composed of a denoising module and super-resolution module is to acquire a high-quality high-resolution face image. The discriminator utilizes some carefully designed loss functions to ensure the quality of the recovered face images. Moreover, we incorporate the identity information and attention mechanism into our network for feasible correlated feature expression and informative feature learning. By jointly performing denoising and face super-resolution, the two tasks can complement each other and attain promising performance. Extensive qualitative and quantitative results show the superiority of our proposed JDSR-GAN over some comparable methods which perform the previous two tasks separately.