arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.16748cs.CVcs.GR
收藏

GANs的双重对比损失和注意力

Dual Contrastive Loss and Attention for GANs

University of Maryland(马里兰大学);Max Planck Institute for Informatics(马克斯·普朗克信息学研究所);NVIDIA(英伟达);Bilkent University(比尔肯特大学);CISPA Helmholtz Center for Information Security(CISPA亥姆霍兹信息安全中心)

Ning Yu, Guilin Liu, Aysegul Dundar, Andrew Tao, Bryan Catanzaro, Larry Davis, Mario Fritz

摘要:

生成对抗网络(GANs)在使用大规模图像数据集时,在无条件图像生成方面产生了令人印象深刻的结果。然而,生成的图像仍然很容易被发现,特别是在具有高方差的数据集上(例如卧室,教堂)。在本文中,我们提出了各种改进措施,以进一步突破图像生成的界限。具体来说,我们提出了一种新的双重对比损失,并表明,这种损失,学习更广义和可区分的表示,以激励一代。此外,我们重新审视了注意力,并在生成器中广泛实验了不同的注意力块。我们发现注意力仍然是成功的图像生成的一个重要模块,即使它没有被用于最近的国家的最先进的模型。最后,我们研究了机器学习中不同的注意力结构,并提出了一种参考的注意力机制。通过结合这些补救措施的优势,我们在几个基准数据集上将最先进的Fréchet Inception Distance(FID)提高了至少17.5%。我们在合成合成场景上获得了更显着的改进(在FID中高达47.5%)。代码和型号可在https://github.com/ningyu1991/AttentionDualContrastGAN上获得。

英文摘要:

Generative Adversarial Networks (GANs) produce impressive results on unconditional image generation when powered with large-scale image datasets. Yet generated images are still easy to spot especially on datasets with high variance (e.g. bedroom, church). In this paper, we propose various improvements to further push the boundaries in image generation. Specifically, we propose a novel dual contrastive loss and show that, with this loss, discriminator learns more generalized and distinguishable representations to incentivize generation. In addition, we revisit attention and extensively experiment with different attention blocks in the generator. We find attention to be still an important module for successful image generation even though it was not used in the recent state-of-the-art models. Lastly, we study different attention architectures in the discriminator, and propose a reference attention mechanism. By combining the strengths of these remedies, we improve the compelling state-of-the-art Fréchet Inception Distance (FID) by at least 17.5% on several benchmark datasets. We obtain even more significant improvements on compositional synthetic scenes (up to 47.5% in FID). Code and models are available at https://github.com/ningyu1991/AttentionDualContrastGAN .

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑