DivCo: Diverse Conditional Image Synthesis via Contrastive Generative Adversarial Network

Li, Rui; Ge, Yixiao; Choi, Ching Lam; Wang, Xiaogang; Li, Hongsheng

doi:10.1109/cvpr46437.2021.01611

Cited by 66 publications

(28 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Due to the constant evolution of GANs during the last few years, these reviews are outdated almost instantaneously. As a result of some relevant and recent researches like [68,176,91] cannot be found in any recent GAN review [3,30]. We consider that a new and more complete review must be done, covering the researches that previous reviews did not fill in and contributing to a deeper and more thorough analysis of the state-of-the-art of GANs.…”

Section: Related Workmentioning

confidence: 99%

A survey on GANs for computer vision: Recent research, analysis and taxonomy

Iglesias¹,

Talavera²,

Díaz-Álvarez³

2022

Preprint

View full text Add to dashboard Cite

In the last few years, there have been several revolutions in the field of deep learning, mainly headlined by the large impact of Generative Adversarial Networks (GANs). GANs not only provide an unique architecture when defining their models, but also generate incredible results which have had a direct impact on society. Due to the significant improvements and new areas of research that GANs have brought, the community is constantly coming up with new researches that make it almost impossible to keep up with the times. Our survey aims to provide a general overview of GANs, showing the latest architectures, optimizations of the loss functions, validation metrics and application areas of the most widely recognized variants. The efficiency of the different variants of the model architecture will be evaluated, as well as showing the best application area; as a vital part of the process, the different metrics for evaluating the performance of GANs and the frequently used loss functions will be analyzed.The final objective of this survey is to provide a summary of the evolution and performance of the GANs which are having better results to guide future researchers in the field.

show abstract

Section: Related Workmentioning

confidence: 99%

A survey on GANs for computer vision: Recent research, analysis and taxonomy

Iglesias¹,

Talavera²,

Díaz-Álvarez³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Take H → N H as an example, given a query image q generated from a latent code, we extract feature representations for generated images, i.e, f = E L (G enc (z, c)). We wish the same factor values c produce the similar image features f , even match with various z, and vice versa [27,34]. Here, we denote the corresponding similar feature as "positive" f + = E L (G enc (z + , c + )) and dissimilar features as "negatives"…”

Section: Adversarial Contrastive Lossmentioning

confidence: 99%

Unpaired Deep Image Dehazing Using Contrastive Disentanglement Learning

Chen¹,

Fan²,

Li³

et al. 2022

Preprint

View full text Add to dashboard Cite

We present an effective unpaired learning based image dehazing network from an unpaired set of clear and hazy images. This paper provides a new perspective to treat image dehazing as a two-class separated factor disentanglement task, i.e, the task-relevant factor of clear image reconstruction and the task-irrelevant factor of haze-relevant distribution. To achieve the disentanglement of these two-class factors in deep feature space, contrastive learning is introduced into a CycleGAN framework to learn disentangled representations by guiding the generated images to be associated with latent factors. With such formulation, the proposed contrastive disentangled dehazing method (CDD-GAN) first develops negative generators to cooperate with the encoder network to update alternately, so as to produce a queue of challenging negative adversaries. Then these negative adversaries are trained end-to-end together with the backbone representation network to enhance the discriminative information and promote factor disentanglement performance by maximizing the adversarial contrastive loss. During the training, we further show that hard negative examples can suppress the task-irrelevant factors and unpaired clear exemples can enhance the task-relevant factors, in order to better facilitate haze removal and help image restoration. Extensive experiments on both synthetic and real-world datasets demonstrate that our method performs favorably against existing stateof-the-art unpaired dehazing approaches.

show abstract

“…Recent years have witnessed the rapid development of image generation supported by a bunch of generative adversarial network (GAN) [9] based methods [1,11,27,29,31,34,35,41,44]. Compared with previous approaches [22,42], GAN-based methods model the domain-specific data distributions better through the specific adversarial training paradigm, i.e., a discriminator is trained to distinguish whether the images are true or false for the optimization of the generator.…”

Section: Related Workmentioning

confidence: 99%

PixelFolder: An Efficient Progressive Pixel Synthesis Network for Image Generation

He¹,

Zhou²,

Zhang³

et al. 2022

Preprint

View full text Add to dashboard Cite

show that with much less expenditure, PixelFolder obtains new state-of-the-art (SOTA) performance on two benchmark datasets, i.e., 3.77 FID and 2.45 FID on FFHQ and LSUN Church, respectively.Meanwhile, PixelFolder is also more efficient than the SOTA methods like StyleGAN2, reducing about 74% computation and 36% parameters, respectively. These results greatly validate the effectiveness of the proposed PixelFolder. Our code is available at https://github.com/BlingHe/PixelFolder.

show abstract

DivCo: Diverse Conditional Image Synthesis via Contrastive Generative Adversarial Network

Cited by 66 publications

References 19 publications

A survey on GANs for computer vision: Recent research, analysis and taxonomy

A survey on GANs for computer vision: Recent research, analysis and taxonomy

Unpaired Deep Image Dehazing Using Contrastive Disentanglement Learning

PixelFolder: An Efficient Progressive Pixel Synthesis Network for Image Generation

Contact Info

Product

Resources

About