Generative Adversarial Networks: A Survey and Taxonomy

By Zhengwei Wang, Qi She, Tomás Ward. 2019

93 citations recorded by the source corpus.

Generative adversarial networks (GANs) have been extensively studied in the past few years. Arguably the revolutionary techniques are in the area of computer vision such as plausible image generation, image to image translation, facial attribute manipulation and similar domains. Despite the significant success achieved in the computer vision field, applying GANs to real-world problems still poses significant challenges, three of which we focus on here: (1) High quality image generation; (2) Diverse image generation; and (3) Stable training. Through an in-depth review of GAN-related research in the literature, we provide an account of the architecture-variants and loss-variants, which have been proposed to handle these three challenges from two perspectives. We propose loss-variants and architecture-variants for classifying the most popular GANs, and discuss the potential improvements with focusing on these two aspects. While several reviews for GANs have been presented to date, none have focused on the review of GAN-variants based on their handling the challenges mentioned above. In this paper, we review and critically discuss 7 architecture-variant GANs and 9 loss-variant GANs for remedying those three challenges. The objective of this review is to provide an insight on the footprint that current GANs research focuses on the performance improvement. Code related to GAN-variants studied in this work is summarized on https://github.com/sheqi/GAN_Review.

Subfields: Computer Vision and Pattern Recognition, Media Technology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Dacheng Tao
2. Junaed Sattar
3. Michael Fulton
4. Jungseok Hong
5. Jing Zhang
6. Zhenan Sun
7. Qi She
8. Jie Gui

In Media Technology

1. Michael Fulton
2. Junaed Sattar
3. Jungseok Hong
4. Dacheng Tao
5. Jieping Ye
6. Yonggang Wen
7. Tomás Ward
8. Qi She

Citations in the corpus

• Self-Supervised Learning: Generative or Contrastive (2020, background): “There exist several comprehensive reviews related to Pre-trained Language Models [103], Generative Adversarial Networks [140], autoencoders, and contrastive learning for visual representation [63].”
• A Generative Approach Towards Improved Robotic Detection of Marine Litter (2019, methodology): “Various generative models [16] have been proposed to generate realistic imagery, and they are used for data augmentation in the non-underwater domain [17] to improve object classification and detection tasks.”
• Generative Adversarial Networks (GANs) (2020, unknown): “References [13] and [14] provided an limited overview of the GANs taxonomy and introduced some of the architecturevariants and loss-variants of GANs.”
• Foundations and Modeling of Dynamic Networks Using Dynamic Graph Neural Networks: A Survey (2020, background): “Generative adverserial networks (GAN) [82] have proven to be very successful in the computer vision field [83].”
• Synthetic-Neuroscore: Using a neuro-AI interface for evaluating generative adversarial networks (2019, background): “There is a growing interest in studying generative adversarial networks (GANs) in the deep learning community [1, 2].”
• A Review on Generative Adversarial Networks: Algorithms, Theory, and Applications (2020, background): “However, MLP does not have good generalization on more complex images [10].”