CLIP-MDGAN: Multi-Discriminator GAN Using CLIP Task Allocation

Shonosuke Gonda, Fumihiko Sakaue, Jun Sato

2025

Abstract

In a Generative Adversarial Network (GAN), in which the generator and discriminator learn adversarially, the performance of the generator can be improved by improving the discriminator’s discriminatory ability. Thus, in this paper, we propose a method to improve the generator’s generative ability by adversarially training a single generator with multiple discriminators, each with different expertise. By each discriminator having different expertise, the overall discriminatory ability of the discriminator is improved, which improves the generator’s performance. However, it is not easy to give multiple discriminators independent expertise. To address this, we propose CLIP-MDGAN, which leverages CLIP, a large-scale learning model that has recently attracted a lot of attention, to classify a dataset into multiple classes with different visual features. Based on CLIP-based classification, each discriminator is assigned a specific subset of images to promote the development of independent expertise. Furthermore, we introduce a method to gradually increase the number of discriminators in adversarial training to reduce instability in training multiple discriminators and reduce training costs.

Paper Citation

in Harvard Style

Gonda S., Sakaue F. and Sato J. (2025). CLIP-MDGAN: Multi-Discriminator GAN Using CLIP Task Allocation. In Proceedings of the 20th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 2: VISAPP; ISBN 978-989-758-728-3, SciTePress, pages 464-470. DOI: 10.5220/0013231900003912

in Bibtex Style

@conference{visapp25,
author={Shonosuke Gonda and Fumihiko Sakaue and Jun Sato},
title={CLIP-MDGAN: Multi-Discriminator GAN Using CLIP Task Allocation},
booktitle={Proceedings of the 20th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 2: VISAPP},
year={2025},
pages={464-470},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013231900003912},
isbn={978-989-758-728-3},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 20th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 2: VISAPP
TI - CLIP-MDGAN: Multi-Discriminator GAN Using CLIP Task Allocation
SN - 978-989-758-728-3
AU - Gonda S.
AU - Sakaue F.
AU - Sato J.
PY - 2025
SP - 464
EP - 470
DO - 10.5220/0013231900003912
PB - SciTePress