3D shape attributes

David F. Fouhey; Abhinav Gupta; Andrew Zisserman

doi:10.1109/CVPR.2016.168

3D shape attributes

David F. Fouhey, Abhinav Gupta, Andrew Zisserman

Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

In this paper we investigate 3D attributes as a means to understand the shape of an object in a single image. To this end, we make a number of contributions: (i) we introduce and define a set of 3D Shape attributes, including planarity, symmetry and occupied space, (ii) we show that such properties can be successfully inferred from a single image using a Convolutional Neural Network (CNN), (iii) we introduce a 143K image dataset of sculptures with 2197 works over 242 artists for training and evaluating the CNN, (iv) we show that the 3D attributes trained on this dataset generalize to images of other (non-sculpture) object classes, and furthermore (v) we show that the CNN also provides a shape embedding that can be used to match previously unseen sculptures largely independent of viewpoint.

Original language	English (US)
Title of host publication	Proceedings - 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016
Publisher	IEEE Computer Society
Pages	1516-1524
Number of pages	9
ISBN (Electronic)	9781467388504
DOIs	https://doi.org/10.1109/CVPR.2016.168
State	Published - Dec 9 2016
Event	29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016 - Las Vegas, United States Duration: Jun 26 2016 → Jul 1 2016

Publication series

Name	Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
Volume	2016-December
ISSN (Print)	1063-6919

Conference

Conference	29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016
Country/Territory	United States
City	Las Vegas
Period	6/26/16 → 7/1/16

ASJC Scopus subject areas

Software
Computer Vision and Pattern Recognition

Access to Document

10.1109/CVPR.2016.168

Cite this

Fouhey, D. F., Gupta, A., & Zisserman, A. (2016). 3D shape attributes. In Proceedings - 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016 (pp. 1516-1524). Article 7780537 (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition; Vol. 2016-December). IEEE Computer Society. https://doi.org/10.1109/CVPR.2016.168

3D shape attributes. / Fouhey, David F.; Gupta, Abhinav; Zisserman, Andrew.
Proceedings - 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016. IEEE Computer Society, 2016. p. 1516-1524 7780537 (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition; Vol. 2016-December).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Fouhey, DF, Gupta, A & Zisserman, A 2016, 3D shape attributes. in Proceedings - 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016., 7780537, Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 2016-December, IEEE Computer Society, pp. 1516-1524, 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016, Las Vegas, United States, 6/26/16. https://doi.org/10.1109/CVPR.2016.168

@inproceedings{ade6e0f23d5745cca60eca1c16bceb8f,

title = "3D shape attributes",

abstract = "In this paper we investigate 3D attributes as a means to understand the shape of an object in a single image. To this end, we make a number of contributions: (i) we introduce and define a set of 3D Shape attributes, including planarity, symmetry and occupied space, (ii) we show that such properties can be successfully inferred from a single image using a Convolutional Neural Network (CNN), (iii) we introduce a 143K image dataset of sculptures with 2197 works over 242 artists for training and evaluating the CNN, (iv) we show that the 3D attributes trained on this dataset generalize to images of other (non-sculpture) object classes, and furthermore (v) we show that the CNN also provides a shape embedding that can be used to match previously unseen sculptures largely independent of viewpoint.",

author = "Fouhey, {David F.} and Abhinav Gupta and Andrew Zisserman",

note = "Publisher Copyright: {\textcopyright} 2016 IEEE.; 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016 ; Conference date: 26-06-2016 Through 01-07-2016",

year = "2016",

month = dec,

day = "9",

doi = "10.1109/CVPR.2016.168",

language = "English (US)",

series = "Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition",

publisher = "IEEE Computer Society",

pages = "1516--1524",

booktitle = "Proceedings - 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016",

}

TY - GEN

T1 - 3D shape attributes

AU - Fouhey, David F.

AU - Gupta, Abhinav

AU - Zisserman, Andrew

PY - 2016/12/9

Y1 - 2016/12/9

N2 - In this paper we investigate 3D attributes as a means to understand the shape of an object in a single image. To this end, we make a number of contributions: (i) we introduce and define a set of 3D Shape attributes, including planarity, symmetry and occupied space, (ii) we show that such properties can be successfully inferred from a single image using a Convolutional Neural Network (CNN), (iii) we introduce a 143K image dataset of sculptures with 2197 works over 242 artists for training and evaluating the CNN, (iv) we show that the 3D attributes trained on this dataset generalize to images of other (non-sculpture) object classes, and furthermore (v) we show that the CNN also provides a shape embedding that can be used to match previously unseen sculptures largely independent of viewpoint.

AB - In this paper we investigate 3D attributes as a means to understand the shape of an object in a single image. To this end, we make a number of contributions: (i) we introduce and define a set of 3D Shape attributes, including planarity, symmetry and occupied space, (ii) we show that such properties can be successfully inferred from a single image using a Convolutional Neural Network (CNN), (iii) we introduce a 143K image dataset of sculptures with 2197 works over 242 artists for training and evaluating the CNN, (iv) we show that the 3D attributes trained on this dataset generalize to images of other (non-sculpture) object classes, and furthermore (v) we show that the CNN also provides a shape embedding that can be used to match previously unseen sculptures largely independent of viewpoint.

UR - http://www.scopus.com/inward/record.url?scp=84986295364&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84986295364&partnerID=8YFLogxK

U2 - 10.1109/CVPR.2016.168

DO - 10.1109/CVPR.2016.168

M3 - Conference contribution

AN - SCOPUS:84986295364

T3 - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition

SP - 1516

EP - 1524

BT - Proceedings - 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016

PB - IEEE Computer Society

T2 - 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016

Y2 - 26 June 2016 through 1 July 2016

ER -

3D shape attributes

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this