Large Scale Metric Learning from Equivalence Constraints

Martin Köstinger, Martin Hirzer, Paul Wohlhart, Peter Roth, Horst Bischof

Publikation: Beitrag in Buch/Bericht/KonferenzbandBeitrag in einem KonferenzbandForschungBegutachtung

Abstract

In this paper, we raise important issues on scalability and the required degree of supervision of existing Mahalanobis metric learning methods. Often rather tedious optimization procedures are applied that become computationally intractable on a large scale. Further, if one considers the constantly growing amount of data it is often infeasible to specify fully supervised labels for all data points. Instead, it is easier to specify labels in form of equivalence constraints. We introduce a simple though effective strategy to learn a distance metric from equivalence constraints, based on a statistical inference perspective. In contrast to existing methods we do not rely on complex optimization problems requiring computationally expensive iterations. Hence, our method is orders of magnitudes faster than comparable methods. Results on a variety of challenging benchmarks with rather diverse nature demonstrate the power of our method. These include faces in unconstrained environments, matching before unseen object instances and person re-identification across spatially disjoint cameras. In the latter two benchmarks we clearly outperform the state-of-the-art.
Originalspracheenglisch
TitelProceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
Herausgeber (Verlag).
Seiten2288-2295
DOIs
PublikationsstatusVeröffentlicht - 2012

Fingerprint

Labels
Scalability
Cameras

Fields of Expertise

  • Information, Communication & Computing

Dies zitieren

Köstinger, M., Hirzer, M., Wohlhart, P., Roth, P., & Bischof, H. (2012). Large Scale Metric Learning from Equivalence Constraints. in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (S. 2288-2295). .. https://doi.org/10.1109/CVPR.2012.6247939

Large Scale Metric Learning from Equivalence Constraints. / Köstinger, Martin; Hirzer, Martin; Wohlhart, Paul; Roth, Peter; Bischof, Horst.

Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). ., 2012. S. 2288-2295.

Publikation: Beitrag in Buch/Bericht/KonferenzbandBeitrag in einem KonferenzbandForschungBegutachtung

Köstinger, M, Hirzer, M, Wohlhart, P, Roth, P & Bischof, H 2012, Large Scale Metric Learning from Equivalence Constraints. in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). ., S. 2288-2295. https://doi.org/10.1109/CVPR.2012.6247939
Köstinger M, Hirzer M, Wohlhart P, Roth P, Bischof H. Large Scale Metric Learning from Equivalence Constraints. in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). . 2012. S. 2288-2295 https://doi.org/10.1109/CVPR.2012.6247939
Köstinger, Martin ; Hirzer, Martin ; Wohlhart, Paul ; Roth, Peter ; Bischof, Horst. / Large Scale Metric Learning from Equivalence Constraints. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). ., 2012. S. 2288-2295
@inproceedings{f51a143b463d4773935f47a28c184f5b,
title = "Large Scale Metric Learning from Equivalence Constraints",
abstract = "In this paper, we raise important issues on scalability and the required degree of supervision of existing Mahalanobis metric learning methods. Often rather tedious optimization procedures are applied that become computationally intractable on a large scale. Further, if one considers the constantly growing amount of data it is often infeasible to specify fully supervised labels for all data points. Instead, it is easier to specify labels in form of equivalence constraints. We introduce a simple though effective strategy to learn a distance metric from equivalence constraints, based on a statistical inference perspective. In contrast to existing methods we do not rely on complex optimization problems requiring computationally expensive iterations. Hence, our method is orders of magnitudes faster than comparable methods. Results on a variety of challenging benchmarks with rather diverse nature demonstrate the power of our method. These include faces in unconstrained environments, matching before unseen object instances and person re-identification across spatially disjoint cameras. In the latter two benchmarks we clearly outperform the state-of-the-art.",
author = "Martin K{\"o}stinger and Martin Hirzer and Paul Wohlhart and Peter Roth and Horst Bischof",
year = "2012",
doi = "10.1109/CVPR.2012.6247939",
language = "English",
pages = "2288--2295",
booktitle = "Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)",
publisher = ".",

}

TY - GEN

T1 - Large Scale Metric Learning from Equivalence Constraints

AU - Köstinger, Martin

AU - Hirzer, Martin

AU - Wohlhart, Paul

AU - Roth, Peter

AU - Bischof, Horst

PY - 2012

Y1 - 2012

N2 - In this paper, we raise important issues on scalability and the required degree of supervision of existing Mahalanobis metric learning methods. Often rather tedious optimization procedures are applied that become computationally intractable on a large scale. Further, if one considers the constantly growing amount of data it is often infeasible to specify fully supervised labels for all data points. Instead, it is easier to specify labels in form of equivalence constraints. We introduce a simple though effective strategy to learn a distance metric from equivalence constraints, based on a statistical inference perspective. In contrast to existing methods we do not rely on complex optimization problems requiring computationally expensive iterations. Hence, our method is orders of magnitudes faster than comparable methods. Results on a variety of challenging benchmarks with rather diverse nature demonstrate the power of our method. These include faces in unconstrained environments, matching before unseen object instances and person re-identification across spatially disjoint cameras. In the latter two benchmarks we clearly outperform the state-of-the-art.

AB - In this paper, we raise important issues on scalability and the required degree of supervision of existing Mahalanobis metric learning methods. Often rather tedious optimization procedures are applied that become computationally intractable on a large scale. Further, if one considers the constantly growing amount of data it is often infeasible to specify fully supervised labels for all data points. Instead, it is easier to specify labels in form of equivalence constraints. We introduce a simple though effective strategy to learn a distance metric from equivalence constraints, based on a statistical inference perspective. In contrast to existing methods we do not rely on complex optimization problems requiring computationally expensive iterations. Hence, our method is orders of magnitudes faster than comparable methods. Results on a variety of challenging benchmarks with rather diverse nature demonstrate the power of our method. These include faces in unconstrained environments, matching before unseen object instances and person re-identification across spatially disjoint cameras. In the latter two benchmarks we clearly outperform the state-of-the-art.

U2 - 10.1109/CVPR.2012.6247939

DO - 10.1109/CVPR.2012.6247939

M3 - Conference contribution

SP - 2288

EP - 2295

BT - Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

PB - .

ER -