Who's Better? Who's Best? Pairwise Deep Ranking for Skill Determination

Hazel Doughty; Dima Damen; Walterio Mayol-Cuevas

doi:10.1109/CVPR.2018.00634

Who's Better? Who's Best? Pairwise Deep Ranking for Skill Determination

Hazel Doughty, Dima Damen, Walterio Mayol-Cuevas

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

117 Citations (Scopus)

327 Downloads (Pure)

Abstract

This paper presents a method for assessing skill from video, applicable to a variety of tasks, ranging from surgery to drawing and rolling pizza dough. We formulate the problem as pairwise (who’s better?) and overall (who’s best?) ranking of video collections, using supervised deep ranking. We propose a novel loss function that learns discriminative features when a pair of videos exhibit variance in skill, and learns shared features when a pair of videos exhibit comparable skill levels. Results demonstrate our method is applicable across tasks, with the percentage of correctly ordered pairs of videos ranging from 70% to 83% for four datasets. We demonstrate the robustness of our approach via sensitivity analysis of its parameters. We see this work as effort toward the automated organization of how-to video collections and overall, generic skill determination in video.

Original language	English
Title of host publication	2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2018)
Subtitle of host publication	Proceedings of a meeting held 18-22 June 2018, Salt Lake City, Utah, USA
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Pages	6057-6066
Number of pages	10
ISBN (Electronic)	9781538664209
ISBN (Print)	9781538664216
DOIs	https://doi.org/10.1109/CVPR.2018.00634
Publication status	Published - Feb 2019
Event	Computer Vision and Pattern Recognition - Salt Lake City, United States Duration: 18 Jun 2018 → 22 Jun 2018

Publication series

Name
ISSN (Print)	1063-6919
ISSN (Electronic)	2575-7075

Conference

Conference	Computer Vision and Pattern Recognition
Abbreviated title	CVPR
Country/Territory	United States
City	Salt Lake City
Period	18/06/18 → 22/06/18

Access to Document

10.1109/CVPR.2018.00634

Full-text PDF (accepted author manuscript)
This is the author accepted manuscript (AAM). The final published version (version of record) is available online via IEEE at https://ieeexplore.ieee.org/document/8578732 . Please refer to any applicable terms of use of the publisher.
Accepted author manuscript, 1.25 MB

Handle.net

Persistent link

Cite this

Doughty, H., Damen, D., & Mayol-Cuevas, W. (2019). Who's Better? Who's Best? Pairwise Deep Ranking for Skill Determination. In 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2018): Proceedings of a meeting held 18-22 June 2018, Salt Lake City, Utah, USA (pp. 6057-6066). Article 8578732 Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/CVPR.2018.00634

@inproceedings{91c0dbee839641049a623aea639a1cad,

title = "Who's Better? Who's Best? Pairwise Deep Ranking for Skill Determination",

abstract = "This paper presents a method for assessing skill from video, applicable to a variety of tasks, ranging from surgery to drawing and rolling pizza dough. We formulate the problem as pairwise (who{\textquoteright}s better?) and overall (who{\textquoteright}s best?) ranking of video collections, using supervised deep ranking. We propose a novel loss function that learns discriminative features when a pair of videos exhibit variance in skill, and learns shared features when a pair of videos exhibit comparable skill levels. Results demonstrate our method is applicable across tasks, with the percentage of correctly ordered pairs of videos ranging from 70% to 83% for four datasets. We demonstrate the robustness of our approach via sensitivity analysis of its parameters. We see this work as effort toward the automated organization of how-to video collections and overall, generic skill determination in video.",

author = "Hazel Doughty and Dima Damen and Walterio Mayol-Cuevas",

year = "2019",

month = feb,

doi = "10.1109/CVPR.2018.00634",

language = "English",

isbn = "9781538664216 ",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

pages = "6057--6066",

booktitle = "2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2018)",

address = "United States",

note = "Computer Vision and Pattern Recognition, CVPR ; Conference date: 18-06-2018 Through 22-06-2018",

}

Doughty, H, Damen, D & Mayol-Cuevas, W 2019, Who's Better? Who's Best? Pairwise Deep Ranking for Skill Determination. in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2018): Proceedings of a meeting held 18-22 June 2018, Salt Lake City, Utah, USA., 8578732, Institute of Electrical and Electronics Engineers (IEEE), pp. 6057-6066, Computer Vision and Pattern Recognition, Salt Lake City, United States, 18/06/18. https://doi.org/10.1109/CVPR.2018.00634

Who's Better? Who's Best? Pairwise Deep Ranking for Skill Determination. / Doughty, Hazel; Damen, Dima ; Mayol-Cuevas, Walterio.
2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2018): Proceedings of a meeting held 18-22 June 2018, Salt Lake City, Utah, USA. Institute of Electrical and Electronics Engineers (IEEE), 2019. p. 6057-6066 8578732.

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

TY - GEN

T1 - Who's Better? Who's Best? Pairwise Deep Ranking for Skill Determination

AU - Doughty, Hazel

AU - Damen, Dima

AU - Mayol-Cuevas, Walterio

PY - 2019/2

Y1 - 2019/2

N2 - This paper presents a method for assessing skill from video, applicable to a variety of tasks, ranging from surgery to drawing and rolling pizza dough. We formulate the problem as pairwise (who’s better?) and overall (who’s best?) ranking of video collections, using supervised deep ranking. We propose a novel loss function that learns discriminative features when a pair of videos exhibit variance in skill, and learns shared features when a pair of videos exhibit comparable skill levels. Results demonstrate our method is applicable across tasks, with the percentage of correctly ordered pairs of videos ranging from 70% to 83% for four datasets. We demonstrate the robustness of our approach via sensitivity analysis of its parameters. We see this work as effort toward the automated organization of how-to video collections and overall, generic skill determination in video.

AB - This paper presents a method for assessing skill from video, applicable to a variety of tasks, ranging from surgery to drawing and rolling pizza dough. We formulate the problem as pairwise (who’s better?) and overall (who’s best?) ranking of video collections, using supervised deep ranking. We propose a novel loss function that learns discriminative features when a pair of videos exhibit variance in skill, and learns shared features when a pair of videos exhibit comparable skill levels. Results demonstrate our method is applicable across tasks, with the percentage of correctly ordered pairs of videos ranging from 70% to 83% for four datasets. We demonstrate the robustness of our approach via sensitivity analysis of its parameters. We see this work as effort toward the automated organization of how-to video collections and overall, generic skill determination in video.

UR - http://www.scopus.com/inward/record.url?scp=85062821845&partnerID=8YFLogxK

U2 - 10.1109/CVPR.2018.00634

DO - 10.1109/CVPR.2018.00634

M3 - Conference Contribution (Conference Proceeding)

SN - 9781538664216

SP - 6057

EP - 6066

BT - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2018)

PB - Institute of Electrical and Electronics Engineers (IEEE)

T2 - Computer Vision and Pattern Recognition

Y2 - 18 June 2018 through 22 June 2018

ER -

Doughty H, Damen D , Mayol-Cuevas W. Who's Better? Who's Best? Pairwise Deep Ranking for Skill Determination. In 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2018): Proceedings of a meeting held 18-22 June 2018, Salt Lake City, Utah, USA. Institute of Electrical and Electronics Engineers (IEEE). 2019. p. 6057-6066. 8578732 Epub 2018 Dec 17. doi: 10.1109/CVPR.2018.00634

Who's Better? Who's Best? Pairwise Deep Ranking for Skill Determination

Abstract

Publication series

Conference

Access to Document

Handle.net

Other files and links

Fingerprint

GLANCE - EPSRC Call for Outlines - User Interaction with ICT

Professor Dima Damen

Cite this

Who's Better? Who's Best? Pairwise Deep Ranking for Skill Determination

Abstract

Publication series

Conference

Access to Document

Handle.net

Other files and links

Fingerprint

Projects

GLANCE - EPSRC Call for Outlines - User Interaction with ICT

Profiles

Professor Dima Damen

Cite this