A latent model of discriminative aspect

Ali Farhadi, Mostafa Kamali Tabrizi, Ian Endres, David Forsyth

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

Abstract

Recognition using appearance features is confounded by phenomena that cause images of the same object to look different, or images of different objects to look the same. This may occur because the same object looks different from different viewing directions, or because two generally different objects have views from which they look similar. In this paper, we introduce the idea of discriminative aspect, a set of latent variables that encode these phenomena. Changes in view direction are one cause of changes in discriminative aspect, but others include changes in texture or lighting. However, images are not labelled with relevant discriminative aspect parameters. We describe a method to improve discrimination by inferring and then using latent discriminative aspect parameters. We apply our method to two parallel problems: object category recognition and human activity recognition. In each case, appearance features are powerful given appropriate training data, but traditionally fail badly under large changes in view. Our method can recognize an object quite reliably in a view for which it possesses no training example. Our method also reweights features to discount accidental similarities in appearance. We demonstrate that our method produces a significant improvement on the state of the art for both object and activity recognition.

Original languageEnglish (US)
Title of host publication2009 IEEE 12th International Conference on Computer Vision, ICCV 2009
Pages948-955
Number of pages8
DOIs
StatePublished - Dec 1 2009
Event12th International Conference on Computer Vision, ICCV 2009 - Kyoto, Japan
Duration: Sep 29 2009Oct 2 2009

Publication series

NameProceedings of the IEEE International Conference on Computer Vision

Other

Other12th International Conference on Computer Vision, ICCV 2009
CountryJapan
CityKyoto
Period9/29/0910/2/09

Fingerprint

Object recognition
Textures
Lighting

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition

Cite this

Farhadi, A., Tabrizi, M. K., Endres, I., & Forsyth, D. (2009). A latent model of discriminative aspect. In 2009 IEEE 12th International Conference on Computer Vision, ICCV 2009 (pp. 948-955). [5459350] (Proceedings of the IEEE International Conference on Computer Vision). https://doi.org/10.1109/ICCV.2009.5459350

A latent model of discriminative aspect. / Farhadi, Ali; Tabrizi, Mostafa Kamali; Endres, Ian; Forsyth, David.

2009 IEEE 12th International Conference on Computer Vision, ICCV 2009. 2009. p. 948-955 5459350 (Proceedings of the IEEE International Conference on Computer Vision).

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

Farhadi, A, Tabrizi, MK, Endres, I & Forsyth, D 2009, A latent model of discriminative aspect. in 2009 IEEE 12th International Conference on Computer Vision, ICCV 2009., 5459350, Proceedings of the IEEE International Conference on Computer Vision, pp. 948-955, 12th International Conference on Computer Vision, ICCV 2009, Kyoto, Japan, 9/29/09. https://doi.org/10.1109/ICCV.2009.5459350
Farhadi A, Tabrizi MK, Endres I, Forsyth D. A latent model of discriminative aspect. In 2009 IEEE 12th International Conference on Computer Vision, ICCV 2009. 2009. p. 948-955. 5459350. (Proceedings of the IEEE International Conference on Computer Vision). https://doi.org/10.1109/ICCV.2009.5459350
Farhadi, Ali ; Tabrizi, Mostafa Kamali ; Endres, Ian ; Forsyth, David. / A latent model of discriminative aspect. 2009 IEEE 12th International Conference on Computer Vision, ICCV 2009. 2009. pp. 948-955 (Proceedings of the IEEE International Conference on Computer Vision).
@inproceedings{ad42ff67c15d403b858bdf3c6ba11dc9,
title = "A latent model of discriminative aspect",
abstract = "Recognition using appearance features is confounded by phenomena that cause images of the same object to look different, or images of different objects to look the same. This may occur because the same object looks different from different viewing directions, or because two generally different objects have views from which they look similar. In this paper, we introduce the idea of discriminative aspect, a set of latent variables that encode these phenomena. Changes in view direction are one cause of changes in discriminative aspect, but others include changes in texture or lighting. However, images are not labelled with relevant discriminative aspect parameters. We describe a method to improve discrimination by inferring and then using latent discriminative aspect parameters. We apply our method to two parallel problems: object category recognition and human activity recognition. In each case, appearance features are powerful given appropriate training data, but traditionally fail badly under large changes in view. Our method can recognize an object quite reliably in a view for which it possesses no training example. Our method also reweights features to discount accidental similarities in appearance. We demonstrate that our method produces a significant improvement on the state of the art for both object and activity recognition.",
author = "Ali Farhadi and Tabrizi, {Mostafa Kamali} and Ian Endres and David Forsyth",
year = "2009",
month = "12",
day = "1",
doi = "10.1109/ICCV.2009.5459350",
language = "English (US)",
isbn = "9781424444205",
series = "Proceedings of the IEEE International Conference on Computer Vision",
pages = "948--955",
booktitle = "2009 IEEE 12th International Conference on Computer Vision, ICCV 2009",

}

TY - GEN

T1 - A latent model of discriminative aspect

AU - Farhadi, Ali

AU - Tabrizi, Mostafa Kamali

AU - Endres, Ian

AU - Forsyth, David

PY - 2009/12/1

Y1 - 2009/12/1

N2 - Recognition using appearance features is confounded by phenomena that cause images of the same object to look different, or images of different objects to look the same. This may occur because the same object looks different from different viewing directions, or because two generally different objects have views from which they look similar. In this paper, we introduce the idea of discriminative aspect, a set of latent variables that encode these phenomena. Changes in view direction are one cause of changes in discriminative aspect, but others include changes in texture or lighting. However, images are not labelled with relevant discriminative aspect parameters. We describe a method to improve discrimination by inferring and then using latent discriminative aspect parameters. We apply our method to two parallel problems: object category recognition and human activity recognition. In each case, appearance features are powerful given appropriate training data, but traditionally fail badly under large changes in view. Our method can recognize an object quite reliably in a view for which it possesses no training example. Our method also reweights features to discount accidental similarities in appearance. We demonstrate that our method produces a significant improvement on the state of the art for both object and activity recognition.

AB - Recognition using appearance features is confounded by phenomena that cause images of the same object to look different, or images of different objects to look the same. This may occur because the same object looks different from different viewing directions, or because two generally different objects have views from which they look similar. In this paper, we introduce the idea of discriminative aspect, a set of latent variables that encode these phenomena. Changes in view direction are one cause of changes in discriminative aspect, but others include changes in texture or lighting. However, images are not labelled with relevant discriminative aspect parameters. We describe a method to improve discrimination by inferring and then using latent discriminative aspect parameters. We apply our method to two parallel problems: object category recognition and human activity recognition. In each case, appearance features are powerful given appropriate training data, but traditionally fail badly under large changes in view. Our method can recognize an object quite reliably in a view for which it possesses no training example. Our method also reweights features to discount accidental similarities in appearance. We demonstrate that our method produces a significant improvement on the state of the art for both object and activity recognition.

UR - http://www.scopus.com/inward/record.url?scp=77953218597&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77953218597&partnerID=8YFLogxK

U2 - 10.1109/ICCV.2009.5459350

DO - 10.1109/ICCV.2009.5459350

M3 - Conference contribution

AN - SCOPUS:77953218597

SN - 9781424444205

T3 - Proceedings of the IEEE International Conference on Computer Vision

SP - 948

EP - 955

BT - 2009 IEEE 12th International Conference on Computer Vision, ICCV 2009

ER -