Visual analogies: A framework for defining aspect categorization

P. Daphne Tsatsoulis; Bryan A. Plummer; David Forsyth

doi:10.1007/978-3-319-49409-8_47

Visual analogies: A framework for defining aspect categorization

P. Daphne Tsatsoulis, Bryan A. Plummer, David Forsyth

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

Abstract

Analogies are common simple word problems (calf is to cow as x is to sheep?) and we use them to identify analogies between images. Let I[A, θ] be an image of object A at view θ. We show how to learn to choose an image I such that I[A, φ] is to I[A, θ] as I is to I[B, θ]. We introduce a framework to identify an image of a familiar object at an unfamiliar angle and extend our method to treat unfamiliar objects. By doing so, we identify pairs of objects that are good at finding new views of one another. This yields an operational notion of aspectual equivalence: objects are equivalent if they can predict each other’s appearance well.

Original language	English (US)
Title of host publication	Computer Vision – ECCV 2016 Workshops, Proceedings
Editors	Gang Hua, Herve Jegou
Publisher	Springer-Verlag
Pages	540-547
Number of pages	8
ISBN (Print)	9783319494081
DOIs	https://doi.org/10.1007/978-3-319-49409-8_47
State	Published - Jan 1 2016
Event	14th European Conference on Computer Vision, ECCV 2016 - Amsterdam, Netherlands Duration: Oct 11 2016 → Oct 14 2016

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	9915 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Other

Other	14th European Conference on Computer Vision, ECCV 2016
Country	Netherlands
City	Amsterdam
Period	10/11/16 → 10/14/16

ASJC Scopus subject areas

Theoretical Computer Science
Computer Science(all)

Access to Document

10.1007/978-3-319-49409-8_47

Fingerprint Dive into the research topics of 'Visual analogies: A framework for defining aspect categorization'. Together they form a unique fingerprint.

Cite this

Tsatsoulis, P. D., Plummer, B. A., & Forsyth, D. (2016). Visual analogies: A framework for defining aspect categorization. In G. Hua, & H. Jegou (Eds.), Computer Vision – ECCV 2016 Workshops, Proceedings (pp. 540-547). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9915 LNCS). Springer-Verlag. https://doi.org/10.1007/978-3-319-49409-8_47

Visual analogies : A framework for defining aspect categorization. / Tsatsoulis, P. Daphne; Plummer, Bryan A.; Forsyth, David.

Computer Vision – ECCV 2016 Workshops, Proceedings. ed. / Gang Hua; Herve Jegou. Springer-Verlag, 2016. p. 540-547 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9915 LNCS).

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

Tsatsoulis, PD, Plummer, BA & Forsyth, D 2016, Visual analogies: A framework for defining aspect categorization. in G Hua & H Jegou (eds), Computer Vision – ECCV 2016 Workshops, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9915 LNCS, Springer-Verlag, pp. 540-547, 14th European Conference on Computer Vision, ECCV 2016, Amsterdam, Netherlands, 10/11/16. https://doi.org/10.1007/978-3-319-49409-8_47

Tsatsoulis PD, Plummer BA, Forsyth D. Visual analogies: A framework for defining aspect categorization. In Hua G, Jegou H, editors, Computer Vision – ECCV 2016 Workshops, Proceedings. Springer-Verlag. 2016. p. 540-547. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-49409-8_47

Tsatsoulis, P. Daphne ; Plummer, Bryan A. ; Forsyth, David. / Visual analogies : A framework for defining aspect categorization. Computer Vision – ECCV 2016 Workshops, Proceedings. editor / Gang Hua ; Herve Jegou. Springer-Verlag, 2016. pp. 540-547 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{18acfbbf87d549a5bc94bdcbee04e2a5,

title = "Visual analogies: A framework for defining aspect categorization",

abstract = "Analogies are common simple word problems (calf is to cow as x is to sheep?) and we use them to identify analogies between images. Let I[A, θ] be an image of object A at view θ. We show how to learn to choose an image I such that I[A, φ] is to I[A, θ] as I is to I[B, θ]. We introduce a framework to identify an image of a familiar object at an unfamiliar angle and extend our method to treat unfamiliar objects. By doing so, we identify pairs of objects that are good at finding new views of one another. This yields an operational notion of aspectual equivalence: objects are equivalent if they can predict each other{\textquoteright}s appearance well.",

author = "Tsatsoulis, {P. Daphne} and Plummer, {Bryan A.} and David Forsyth",

year = "2016",

month = jan,

day = "1",

doi = "10.1007/978-3-319-49409-8_47",

language = "English (US)",

isbn = "9783319494081",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer-Verlag",

pages = "540--547",

editor = "Gang Hua and Herve Jegou",

booktitle = "Computer Vision – ECCV 2016 Workshops, Proceedings",

note = "14th European Conference on Computer Vision, ECCV 2016 ; Conference date: 11-10-2016 Through 14-10-2016",

}

TY - GEN

T1 - Visual analogies

T2 - 14th European Conference on Computer Vision, ECCV 2016

AU - Tsatsoulis, P. Daphne

AU - Plummer, Bryan A.

AU - Forsyth, David

PY - 2016/1/1

Y1 - 2016/1/1

N2 - Analogies are common simple word problems (calf is to cow as x is to sheep?) and we use them to identify analogies between images. Let I[A, θ] be an image of object A at view θ. We show how to learn to choose an image I such that I[A, φ] is to I[A, θ] as I is to I[B, θ]. We introduce a framework to identify an image of a familiar object at an unfamiliar angle and extend our method to treat unfamiliar objects. By doing so, we identify pairs of objects that are good at finding new views of one another. This yields an operational notion of aspectual equivalence: objects are equivalent if they can predict each other’s appearance well.

AB - Analogies are common simple word problems (calf is to cow as x is to sheep?) and we use them to identify analogies between images. Let I[A, θ] be an image of object A at view θ. We show how to learn to choose an image I such that I[A, φ] is to I[A, θ] as I is to I[B, θ]. We introduce a framework to identify an image of a familiar object at an unfamiliar angle and extend our method to treat unfamiliar objects. By doing so, we identify pairs of objects that are good at finding new views of one another. This yields an operational notion of aspectual equivalence: objects are equivalent if they can predict each other’s appearance well.

UR - http://www.scopus.com/inward/record.url?scp=85006067156&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85006067156&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-49409-8_47

DO - 10.1007/978-3-319-49409-8_47

M3 - Conference contribution

AN - SCOPUS:85006067156

SN - 9783319494081

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 540

EP - 547

BT - Computer Vision – ECCV 2016 Workshops, Proceedings

A2 - Hua, Gang

A2 - Jegou, Herve

PB - Springer-Verlag

Y2 - 11 October 2016 through 14 October 2016

ER -