Single Image 3D Without a Single 3D Image

David F. Fouhey; Wajahat Hussain; Abhinav Gupta; Martial Hebert

ICCV 2015poster28 citations

Single Image 3D Without a Single 3D Image

David F. Fouhey, Wajahat Hussain, Abhinav Gupta, Martial Hebert

Abstract

Do we really need 3D labels in order to learn how to predict 3D? In this paper, we show that one can learn a mapping from appearance to 3D properties without ever seeing a single explicit 3D label. Rather than use explicit supervision, we use the regularity of indoor scenes to learn the mapping in a completely unsupervised manner. We demonstrate this on both a standard 3D scene understanding dataset as well as Internet images for which 3D is unavailable, precluding supervised learning. Despite never seeing a 3D label, our method produces competitive results.

BibTeX

@inproceedings{iccv2015_singleimage3dwit,
  title = {Single Image 3D Without a Single 3D Image},
  author = {David F. Fouhey and Wajahat Hussain and Abhinav Gupta and Martial Hebert},
  booktitle = {ICCV 2015},
  year = {2015}
}