Symmetry reCAPTCHA

Christopher Funk, Yanxi Liu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

12 Citations (Scopus)

Abstract

This paper is a reaction to the poor performance of symmetry detection algorithms on real-world images, benchmarked since CVPR 2011. Our systematic study reveals significant difference between human labeled (reflection and rotation) symmetries on photos and the output of computer vision algorithms on the same photo set. We exploit this human-machine symmetry perception gap by proposing a novel symmetry-based Turing test. By leveraging a comprehensive user interface, we collected more than 78,000 symmetry labels from 400 Amazon Mechanical Turk raters on 1,200 photos from the Microsoft COCO dataset. Using a set of ground-truth symmetries automatically generated from noisy human labels, the effectiveness of our work is evidenced by a separate test where over 96% success rate is achieved. We demonstrate statistically significant outcomes for using symmetry perception as a powerful, alternative, image-based reCAPTCHA.

Original languageEnglish (US)
Title of host publicationProceedings - 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016
PublisherIEEE Computer Society
Pages5165-5174
Number of pages10
ISBN (Electronic)9781467388504
DOIs
StatePublished - Dec 9 2016
Event29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016 - Las Vegas, United States
Duration: Jun 26 2016Jul 1 2016

Publication series

NameProceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
Volume2016-December
ISSN (Print)1063-6919

Conference

Conference29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016
CountryUnited States
CityLas Vegas
Period6/26/167/1/16

Fingerprint

Labels
Computer vision
User interfaces

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Vision and Pattern Recognition

Cite this

Funk, C., & Liu, Y. (2016). Symmetry reCAPTCHA. In Proceedings - 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016 (pp. 5165-5174). [7780927] (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.2016.558
Funk, Christopher ; Liu, Yanxi. / Symmetry reCAPTCHA. Proceedings - 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016. IEEE Computer Society, 2016. pp. 5165-5174 (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition).
@inproceedings{2c0abc3d81a14eb994ae904007807a5d,
title = "Symmetry reCAPTCHA",
abstract = "This paper is a reaction to the poor performance of symmetry detection algorithms on real-world images, benchmarked since CVPR 2011. Our systematic study reveals significant difference between human labeled (reflection and rotation) symmetries on photos and the output of computer vision algorithms on the same photo set. We exploit this human-machine symmetry perception gap by proposing a novel symmetry-based Turing test. By leveraging a comprehensive user interface, we collected more than 78,000 symmetry labels from 400 Amazon Mechanical Turk raters on 1,200 photos from the Microsoft COCO dataset. Using a set of ground-truth symmetries automatically generated from noisy human labels, the effectiveness of our work is evidenced by a separate test where over 96{\%} success rate is achieved. We demonstrate statistically significant outcomes for using symmetry perception as a powerful, alternative, image-based reCAPTCHA.",
author = "Christopher Funk and Yanxi Liu",
year = "2016",
month = "12",
day = "9",
doi = "10.1109/CVPR.2016.558",
language = "English (US)",
series = "Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition",
publisher = "IEEE Computer Society",
pages = "5165--5174",
booktitle = "Proceedings - 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016",
address = "United States",

}

Funk, C & Liu, Y 2016, Symmetry reCAPTCHA. in Proceedings - 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016., 7780927, Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 2016-December, IEEE Computer Society, pp. 5165-5174, 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016, Las Vegas, United States, 6/26/16. https://doi.org/10.1109/CVPR.2016.558

Symmetry reCAPTCHA. / Funk, Christopher; Liu, Yanxi.

Proceedings - 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016. IEEE Computer Society, 2016. p. 5165-5174 7780927 (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition; Vol. 2016-December).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Symmetry reCAPTCHA

AU - Funk, Christopher

AU - Liu, Yanxi

PY - 2016/12/9

Y1 - 2016/12/9

N2 - This paper is a reaction to the poor performance of symmetry detection algorithms on real-world images, benchmarked since CVPR 2011. Our systematic study reveals significant difference between human labeled (reflection and rotation) symmetries on photos and the output of computer vision algorithms on the same photo set. We exploit this human-machine symmetry perception gap by proposing a novel symmetry-based Turing test. By leveraging a comprehensive user interface, we collected more than 78,000 symmetry labels from 400 Amazon Mechanical Turk raters on 1,200 photos from the Microsoft COCO dataset. Using a set of ground-truth symmetries automatically generated from noisy human labels, the effectiveness of our work is evidenced by a separate test where over 96% success rate is achieved. We demonstrate statistically significant outcomes for using symmetry perception as a powerful, alternative, image-based reCAPTCHA.

AB - This paper is a reaction to the poor performance of symmetry detection algorithms on real-world images, benchmarked since CVPR 2011. Our systematic study reveals significant difference between human labeled (reflection and rotation) symmetries on photos and the output of computer vision algorithms on the same photo set. We exploit this human-machine symmetry perception gap by proposing a novel symmetry-based Turing test. By leveraging a comprehensive user interface, we collected more than 78,000 symmetry labels from 400 Amazon Mechanical Turk raters on 1,200 photos from the Microsoft COCO dataset. Using a set of ground-truth symmetries automatically generated from noisy human labels, the effectiveness of our work is evidenced by a separate test where over 96% success rate is achieved. We demonstrate statistically significant outcomes for using symmetry perception as a powerful, alternative, image-based reCAPTCHA.

UR - http://www.scopus.com/inward/record.url?scp=84986247596&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84986247596&partnerID=8YFLogxK

U2 - 10.1109/CVPR.2016.558

DO - 10.1109/CVPR.2016.558

M3 - Conference contribution

AN - SCOPUS:84986247596

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

SP - 5165

EP - 5174

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

PB - IEEE Computer Society

ER -

Funk C, Liu Y. Symmetry reCAPTCHA. In Proceedings - 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016. IEEE Computer Society. 2016. p. 5165-5174. 7780927. (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition). https://doi.org/10.1109/CVPR.2016.558