A Multi-Task Learning Framework for Image Restoration Using a Novel Generative Adversarial Network

Rim Walha, Rim Walha, Fadoua Drira, Rania Bedhief

2024

Abstract

In the last years, deep learning has gained growing popularity in image restoration, becoming the efficient mainstream for the subsequent higher level computer vision processing tasks. In particular, image restoration is a challenging task due to the high variations of degradations faced in the real-world scenarios. In this study, we introduce an efficient multi-task generative adversarial learning based framework as a practical solution suitable for various types of image degradations. We apply recent advancements in deep learning to design, build and train such a framework that can deal with several image restoration tasks treated simultaneously. More precisely, the main specificities of the proposed architecture are: (1) the introduction of a novel generator based on an encoder with separate decoders, (2) the utilization of low-level multi-scale features within the encoder component of our architecture, (3) the incorporation of the multi-scale transformer technique in each decoder in order to learn and share the low-level features representations among different tasks. Our experimental study demonstrates the efficiency and the robustness of the proposed framework for two specific image restoration tasks including image deblurring and image denoising. Moreover, it achieves high performance results that exceed those of state-of-the-art methods when evaluated on the same datasets.

Paper Citation

in Harvard Style

Walha R., Drira F. and Bedhief R. (2024). A Multi-Task Learning Framework for Image Restoration Using a Novel Generative Adversarial Network. In Proceedings of the 16th International Conference on Agents and Artificial Intelligence - Volume 3: ICAART; ISBN 978-989-758-680-4, SciTePress, pages 928-935. DOI: 10.5220/0012421000003636

in Bibtex Style

@conference{icaart24,
author={Rim Walha and Fadoua Drira and Rania Bedhief},
title={A Multi-Task Learning Framework for Image Restoration Using a Novel Generative Adversarial Network},
booktitle={Proceedings of the 16th International Conference on Agents and Artificial Intelligence - Volume 3: ICAART},
year={2024},
pages={928-935},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012421000003636},
isbn={978-989-758-680-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 16th International Conference on Agents and Artificial Intelligence - Volume 3: ICAART
TI - A Multi-Task Learning Framework for Image Restoration Using a Novel Generative Adversarial Network
SN - 978-989-758-680-4
AU - Walha R.
AU - Drira F.
AU - Bedhief R.
PY - 2024
SP - 928
EP - 935
DO - 10.5220/0012421000003636
PB - SciTePress