当前位置: X-MOL 学术IEEE Trans. Neural Netw. Learn. Syst. › 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.)
Bayesian Cycle-Consistent Generative Adversarial Networks via Marginalizing Latent Sampling
IEEE Transactions on Neural Networks and Learning Systems ( IF 10.2 ) Pub Date : 2020-09-03 , DOI: 10.1109/tnnls.2020.3017669
Haoran You , Yu Cheng , Tianheng Cheng , Chunliang Li , Pan Zhou

Recent techniques built on generative adversarial networks (GANs), such as cycle-consistent GANs, are able to learn mappings among different domains built from unpaired data sets, through min–max optimization games between generators and discriminators. However, it remains challenging to stabilize the training process and thus cyclic models fall into mode collapse accompanied by the success of discriminator. To address this problem, we propose an novel Bayesian cyclic model and an integrated cyclic framework for interdomain mappings. The proposed method motivated by Bayesian GAN explores the full posteriors of cyclic model via sampling latent variables and optimizes the model with maximum a posteriori (MAP) estimation. Hence, we name it Bayesian CycleGAN. In addition, original CycleGAN cannot generate diversified results. But it is feasible for Bayesian framework to diversify generated images by replacing restricted latent variables in inference process. We evaluate the proposed Bayesian CycleGAN on multiple benchmark data sets, including Cityscapes, Maps, and Monet2photo. The proposed method improve the per-pixel accuracy by 15% for the Cityscapes semantic segmentation task within origin framework and improve 20% within the proposed integrated framework, showing better resilience to imbalance confrontation. The diversified results of Monet2Photo style transfer also demonstrate its superiority over original cyclic model. We provide codes for all of our experiments in https://github.com/ranery/Bayesian-CycleGAN .

中文翻译:

通过边缘化潜在采样的贝叶斯循环一致生成对抗网络

最近建立在生成对抗网络 (GAN) 上的技术,例如循环一致的 GAN,能够通过生成器和鉴别器之间的最小-最大优化博弈,学习从不成对数据集构建的不同域之间的映射。然而,稳定训练过程仍然具有挑战性,因此循环模型随着鉴别器的成功陷入模式崩溃。为了解决这个问题,我们提出了一种新的贝叶斯循环模型和一个用于域间映射的集成循环框架。所提出的由贝叶斯 GAN 启发的方法通过采样潜在变量探索循环模型的完整后验,并以最大后验 (MAP) 估计。因此,我们将其命名为贝叶斯 CycleGAN。此外,原始的 CycleGAN 无法产生多样化的结果。但是贝叶斯框架通过在推理过程中替换受限制的潜在变量来使生成的图像多样化是可行的。我们在多个基准数据集上评估了提议的贝叶斯 CycleGAN,包括 Cityscapes、Maps 和 Monet2photo。所提出的方法在原始框架内将 Cityscapes 语义分割任务的每像素精度提高了 15%,并在所提出的集成框架内提高了 20%,显示出更好的对不平衡对抗的弹性。Monet2Photo 风格迁移的多样化结果也证明了其优于原始循环模型。我们为我们所有的实验提供代码https://github.com/ranery/Bayesian-CycleGAN .
更新日期:2020-09-03
down
wechat
bug