arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2006.04026cs.CV
收藏

SharinGAN:结合合成和真实数据进行无监督几何估计

SharinGAN: Combining Synthetic and Real Data for Unsupervised Geometry Estimation

University of Maryland(马里兰大学);Amazon AWS(亚马逊云科技)

Koutilya PNVR, Hao Zhou, David Jacobs

摘要:

我们提出了一种新的方法,用于在训练网络时将合成图像和真实图像相结合,以从单个图像中确定几何信息。我们提出了一种方法,将这两种图像类型映射到一个单一的,共享的域。这是连接到一个主要的网络进行端到端的培训。理想情况下,这会导致来自两个域的映像向主网络提供共享信息。我们的实验表明,显着的改进,在两个重要的领域,表面正常估计的人脸和单眼深度估计的户外场景,无论是在无监督的设置的最先进的。

英文摘要:

We propose a novel method for combining synthetic and real images when training networks to determine geometric information from a single image. We suggest a method for mapping both image types into a single, shared domain. This is connected to a primary network for end-to-end training. Ideally, this results in images from two domains that present shared information to the primary network. Our experiments demonstrate significant improvements over the state-of-the-art in two important domains, surface normal estimation of human faces and monocular depth estimation for outdoor scenes, both in an unsupervised setting.

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。

↑