首页> 外文期刊>IEEE Transactions on Pattern Analysis and Machine Intelligence >StackGAN++: Realistic Image Synthesis with Stacked Generative Adversarial Networks
【24h】

StackGAN++: Realistic Image Synthesis with Stacked Generative Adversarial Networks

机译:StackGAN ++:具有堆叠式生成对抗网络的逼真的图像合成

获取原文
获取原文并翻译 | 示例

摘要

Although Generative Adversarial Networks (GANs) have shown remarkable success in various tasks, they still face challenges in generating high quality images. In this paper, we propose Stacked Generative Adversarial Networks (StackGANs) aimed at generating high-resolution photo-realistic images. First, we propose a two-stage generative adversarial network architecture, StackGAN-v1, for text-to-image synthesis. The Stage-I GAN sketches the primitive shape and colors of a scene based on a given text description, yielding low-resolution images. The Stage-II GAN takes Stage-I results and the text description as inputs, and generates high-resolution images with photo-realistic details. Second, an advanced multi-stage generative adversarial network architecture, StackGAN-v2, is proposed for both conditional and unconditional generative tasks. Our StackGAN-v2 consists of multiple generators and multiple discriminators arranged in a tree-like structure; images at multiple scales corresponding to the same scene are generated from different branches of the tree. StackGAN-v2 shows more stable training behavior than StackGAN-v1 by jointly approximating multiple distributions. Extensive experiments demonstrate that the proposed stacked generative adversarial networks significantly outperform other state-of-the-art methods in generating photo-realistic images.
机译:尽管生成对抗网络(GAN)在各种任务中均显示出了非凡的成功,但它们在生成高质量图像方面仍然面临挑战。在本文中,我们提出了旨在产生高分辨率照片级逼真的图像的堆栈式生成对抗网络(StackGANs)。首先,我们提出了一个两阶段的生成对抗网络架构StackGAN-v1,用于文本到图像的合成。 Stage-I GAN根据给定的文本描述来绘制场景的原始形状和颜色,从而生成低分辨率图像。 Stage-II GAN将Stage-I的结果和文本描述作为输入,并生成具有逼真的细节的高分辨率图像。其次,针对条件和非条件生成任务,提出了一种先进的多阶段生成对抗网络架构StackGAN-v2。我们的StackGAN-v2由多个生成器和多个鉴别器以树状结构排列;从树的不同分支生成对应于同一场景的多个比例的图像。通过共同逼近多个分布,StackGAN-v2显示出比StackGAN-v1更稳定的训练行为。大量实验表明,所提出的堆叠式生成对抗网络在生成照片级逼真的图像方面明显优于其他最新技术。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号