TY - GEN
T1 - Single image 3D without a single 3D image
AU - Fouhey, David F.
AU - Hussain, Wajahat
AU - Gupta, Abhinav
AU - Hebert, Martial
N1 - Publisher Copyright:
© 2015 IEEE.
PY - 2015/2/17
Y1 - 2015/2/17
N2 - 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.
AB - 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.
UR - http://www.scopus.com/inward/record.url?scp=84973888903&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84973888903&partnerID=8YFLogxK
U2 - 10.1109/ICCV.2015.126
DO - 10.1109/ICCV.2015.126
M3 - Conference contribution
AN - SCOPUS:84973888903
T3 - Proceedings of the IEEE International Conference on Computer Vision
SP - 1053
EP - 1061
BT - 2015 International Conference on Computer Vision, ICCV 2015
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 15th IEEE International Conference on Computer Vision, ICCV 2015
Y2 - 11 December 2015 through 18 December 2015
ER -