EUVIMEDEUVIMED entdecken
Crossref

Multiobjective Evolutionary Generative Adversarial Network Compression for Image Translation

Yao Zhou, Bing Hu, Xianglei Yuan, Kaide Huang, Zhang Yi, Gary G. Yen

IEEE Transactions on Evolutionary Computation · 2024-06 · 28 (3) · 798-809

DOI: 10.1109/tevc.2023.3261135 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Evolutionary Computation, 2024-06. Verfasst von Yao Zhou und 5 weiteren Autor:innen. Die Arbeit wurde bisher 24-mal zitiert (Crossref) und nennt 61 Literaturangaben. Gefördert von: National Natural Science Foundation of China, National Key Research and Development Program of China, National Postdoctoral Program for Innovative Talents.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Generative adversarial networks (GANs) have achieved remarkable success in image translation tasks. However, its prohibitive computational overhead has been a major hurdle for deployment on resource-constrained platforms. Due to the training instability and the complicated network architecture, existing model acceleration techniques cannot appropriately handle the GAN compression problem. To cope with these difficulties, we propose a multiobjective evolutionary algorithm to compress GAN models in the context of image translation tasks, which is termed MEGC. Particularly, the conflict between the computational cost of the GAN model and the quality of the generated image is explicitly modeled as a two-objective optimization problem, and the evolved Pareto set is utilized to guide the sampling process during supernet training, which can in turn divert the focus of the supernet training to well-performing compact subnets. Besides, an evaluation-free strategy is introduced to facilitate exploration in the search space while incurring no extra computational cost. Based on the above design, the proposed MEGC eliminates the requirement of subnet searching in the post-processing procedure. Experiments on image translation tasks under paired and unpaired settings demonstrate the effectiveness of the proposed MEGC on reducing the computational cost of GANs while improving the quality of generated images compared to those of the full models.

Förderung

  • National Natural Science Foundation of China
  • National Key Research and Development Program of China
  • National Postdoctoral Program for Innovative Talents
  • Sichuan Science and Technology Planning Project

Zitieren

APA

Zhou, Y., Hu, B., Yuan, X., Huang, K., Yi, Z., & Yen, G. G. (2024). Multiobjective Evolutionary Generative Adversarial Network Compression for Image Translation. IEEE Transactions on Evolutionary Computation, 28(3), 798-809. https://doi.org/10.1109/tevc.2023.3261135

Vancouver

Zhou Y, Hu B, Yuan X, Huang K, Yi Z, Yen GG. Multiobjective Evolutionary Generative Adversarial Network Compression for Image Translation. IEEE Transactions on Evolutionary Computation. 2024;28(3):798-809. doi:10.1109/tevc.2023.3261135

BibTeX
@article{Zhou2024,
  author = {Zhou, Yao and Hu, Bing and Yuan, Xianglei and Huang, Kaide and Yi, Zhang and Yen, Gary G.},
  title = {Multiobjective Evolutionary Generative Adversarial Network Compression for Image Translation},
  journal = {IEEE Transactions on Evolutionary Computation},
  year = {2024},
  volume = {28},
  number = {3},
  pages = {798--809},
  doi = {10.1109/tevc.2023.3261135}
}
Kostenlos starten →