arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1806.09573cs.CV
收藏

使用质量评估网络从视频中学习单图像深度

Learning Single-Image Depth from Videos using Quality Assessment Networks

University of Michigan, Ann Arbor(密歇根大学安娜堡分校);Princeton University(普林斯顿大学)

Weifeng Chen, Shengyi Qian, Jia Deng

摘要:

在野外从单个图像进行深度估计仍然是一个具有挑战性的问题。一个主要障碍是缺乏高质量的野外图像训练数据。在本文中,我们提出了一种方法来自动生成这样的数据,通过结构从运动(SfM)在互联网上的视频。该方法的核心是一个质量评估网络,它可以识别从SfM获得的高质量重建。使用这种方法,我们从大量的YouTube视频中收集单视图深度训练数据,并构建一个新的数据集,称为YouTube3D。实验表明,YouTube3D在训练深度估计网络方面是有效的,并在单视图深度估计领域取得了一定的进展。

英文摘要:

Depth estimation from a single image in the wild remains a challenging problem. One main obstacle is the lack of high-quality training data for images in the wild. In this paper we propose a method to automatically generate such data through Structure-from-Motion (SfM) on Internet videos. The core of this method is a Quality Assessment Network that identifies high-quality reconstructions obtained from SfM. Using this method, we collect single-view depth training data from a large number of YouTube videos and construct a new dataset called YouTube3D. Experiments show that YouTube3D is useful in training depth estimation networks and advances the state of the art of single-view depth estimation in the wild.

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。

↑