arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2006.10705cs.LGcs.CVstat.ML
收藏

Set Distribution Networks:一个图像集生成模型

Set Distribution Networks: a Generative Model for Sets of Images

Apple Inc.(苹果公司)

Shuangfei Zhai, Walter Talbott, Miguel Angel Bautista, Carlos Guestrin, Josh M. Susskind

摘要:

具有共享特征的图像自然形成集合。例如,在面部验证基准中,相同身份的图像形成集合。对于生成模型,处理集合的标准方法是将每个集合表示为一个热向量,并学习条件生成模型$p(\mathbf{x}|\mathbf{y})$。这种表示假设集合的数量是有限的和已知的,使得集合上的分布简化为简单的多项分布。相比之下,我们研究一个更通用的问题,其中集合的数量是大的和未知的。我们介绍了集合分布网络(SDNs),这是一个学习自动编码和自由生成集合的新框架。我们通过联合学习集合编码器,集合编码器,集合生成器和集合先验来实现这一点。我们表明,SDNs能够重建图像集,保留我们的基准数据集中输入的显着属性,并且还能够生成新的对象/身份。我们分别使用预训练的3D重建网络和人脸验证网络来检查SDN生成的集合,作为评估生成的图像集质量的一种新方法。

英文摘要:

Images with shared characteristics naturally form sets. For example, in a face verification benchmark, images of the same identity form sets. For generative models, the standard way of dealing with sets is to represent each as a one hot vector, and learn a conditional generative model $p(\mathbf{x}|\mathbf{y})$. This representation assumes that the number of sets is limited and known, such that the distribution over sets reduces to a simple multinomial distribution. In contrast, we study a more generic problem where the number of sets is large and unknown. We introduce Set Distribution Networks (SDNs), a novel framework that learns to autoencode and freely generate sets. We achieve this by jointly learning a set encoder, set discriminator, set generator, and set prior. We show that SDNs are able to reconstruct image sets that preserve salient attributes of the inputs in our benchmark datasets, and are also able to generate novel objects/identities. We examine the sets generated by SDN with a pre-trained 3D reconstruction network and a face verification network, respectively, as a novel way to evaluate the quality of generated sets of images.

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。

↑