arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2212.03361cs.LGcs.CV
收藏

基于潜空间映射的域转换

Domain Translation via Latent Space Mapping

INSA Rouen Normandie(诺曼底鲁昂国立应用科学学院);Univ Rouen Normandie(诺曼底鲁昂大学);LITIS UR 4108(LITIS 4108联合研究单位)

Tsiry Mayet, Simon Bernard, Clement Chatelain, Romain Herault

摘要:

在本文中,我们研究多域转换的问题:给定域$A$的一个元素$a$,我们希望在另一个域$B$中生成一个相应的$b$样本,反之亦然。在多个域中获取监督可能是一项繁琐的任务,我们还建议在监督作为一对$(a,b)\sim A\times B$可用时学习从一个域到另一个域的这种转换,并在只有$a\sim A$或只有$b\sim B$可用时利用可能的未配对数据。我们引入了一个新的统一的框架,称为潜在空间映射(\model),利用流形假设,以学习,从每个域,一个潜在的空间。与现有的方法不同,我们建议通过学习域对之间的每个依赖关系,使用可用域进一步正则化每个潜在空间。我们在三个任务中评估我们的方法,执行i)具有图像翻译的合成数据集,ii)医学图像的语义分割的现实任务,以及iii)面部标志检测的现实任务。

英文摘要:

In this paper, we investigate the problem of multi-domain translation: given an element $a$ of domain $A$, we would like to generate a corresponding $b$ sample in another domain $B$, and vice versa. Acquiring supervision in multiple domains can be a tedious task, also we propose to learn this translation from one domain to another when supervision is available as a pair $(a,b)\sim A\times B$ and leveraging possible unpaired data when only $a\sim A$ or only $b\sim B$ is available. We introduce a new unified framework called Latent Space Mapping (\model) that exploits the manifold assumption in order to learn, from each domain, a latent space. Unlike existing approaches, we propose to further regularize each latent space using available domains by learning each dependency between pairs of domains. We evaluate our approach in three tasks performing i) synthetic dataset with image translation, ii) real-world task of semantic segmentation for medical images, and iii) real-world task of facial landmark detection.

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。

↑