当前位置: X-MOL 学术Pattern Recogn. Lett. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Fast transformation of discriminators into encoders using pre-trained GANs
Pattern Recognition Letters ( IF 5.1 ) Pub Date : 2021-11-29 , DOI: 10.1016/j.patrec.2021.11.026
Cheng Yu 1, 2 , Wenmin Wang 3, 4
Affiliation  

A typical generative adversarial network (GAN) consists of a generator and a discriminator. Currently, finely tuned deep GANs can synthesize high-quality (HQ) images via their generators. However, the discriminator in typical GANs is only able to distinguish true or fake images in the training process. Moreover, some synthesized images from GANs are imperfect, and we can not reconstruct images via GANs. In this paper, we revisit pre-trained GANs and offer a self-supervised method to quickly transform GAN’s discriminators into encoders. We reuse parameters of the GAN’s discriminator and replace its output layer, so it can be transformed into an encoder and output reformed latent vectors. The transformation makes GAN architecture more symmetrical and allows for better performance. Based on the method, GANs can be made to reconstruct synthesized images via GAN encoders. Compared to synthesized images, these reconstructions can maintain or even attain higher quality. The code and pre-trained models are available at https://github.com/disanda/GAN-Encoder-Sym.



中文翻译:

使用预训练的 GAN 将鉴别器快速转换为编码器

典型的生成对抗网络 (GAN) 由生成器和鉴别器组成。目前,微调的深度 GAN 可以通过其生成器合成高质量 (HQ) 图像。然而,典型的 GAN 中的鉴别器只能在训练过程中区分真假图像。此外,一些来自 GAN 的合成图像是不完美的,我们无法通过 GAN 重建图像。在本文中,我们重新审视了预训练的 GAN,并提供了一种自我监督的方法来快速将 GAN 的鉴别器转换为编码器。我们重用 GAN 鉴别器的参数并替换其输出层,因此可以将其转换为编码器并输出经过改造的潜在向量。这种转换使 GAN 架构更加对称并允许更好的性能。基于该方法,GAN 可以通过 GAN 编码器重建合成图像。与合成图像相比,这些重建可以保持甚至获得更高的质量。代码和预训练模型可从 https://github.com/disanda/GAN-Encoder-Sym 获得。

更新日期:2021-12-14
down
wechat
bug