Hierarchical Gaussian Descriptor for Person Re-identification

Tetsu Matsukawa, Takahiro Okabe, Einoshin Suzuki, Yoichi Sato

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

289 Citations (Scopus)

Abstract

Describing the color and textural information of a person image is one of the most crucial aspects of person re-identification. In this paper, we present a novel descriptor based on a hierarchical distribution of pixel features. A hierarchical covariance descriptor has been successfully applied for image classification. However, the mean information of pixel features, which is absent in covariance, tends to be major discriminative information of person images. To solve this problem, we describe a local region in an image via hierarchical Gaussian distribution in which both means and covariances are included in their parameters. More specifically, we model the region as a set of multiple Gaussian distributions in which each Gaussian represents the appearance of a local patch. The characteristics of the set of Gaussians are again described by another Gaussian distribution. In both steps, unlike the hierarchical covariance descriptor, the proposed descriptor can model both the mean and the covariance information of pixel features properly. The results of experiments conducted on five databases indicate that the proposed descriptor exhibits re-markably high performance which outperforms the state-of-the-art descriptors for person re-identification.

Original languageEnglish
Title of host publicationProceedings - 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016
PublisherIEEE Computer Society
Pages1363-1372
Number of pages10
ISBN (Electronic)9781467388504
DOIs
Publication statusPublished - 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

Gaussian distribution
Pixels
Image classification
Color
Experiments

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Vision and Pattern Recognition

Cite this

Matsukawa, T., Okabe, T., Suzuki, E., & Sato, Y. (2016). Hierarchical Gaussian Descriptor for Person Re-identification. In Proceedings - 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016 (pp. 1363-1372). [7780521] (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.2016.152

Hierarchical Gaussian Descriptor for Person Re-identification. / Matsukawa, Tetsu; Okabe, Takahiro; Suzuki, Einoshin; Sato, Yoichi.

Proceedings - 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016. IEEE Computer Society, 2016. p. 1363-1372 7780521 (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

Matsukawa, T, Okabe, T, Suzuki, E & Sato, Y 2016, Hierarchical Gaussian Descriptor for Person Re-identification. in Proceedings - 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016., 7780521, Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 2016-December, IEEE Computer Society, pp. 1363-1372, 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016, Las Vegas, United States, 6/26/16. https://doi.org/10.1109/CVPR.2016.152
Matsukawa T, Okabe T, Suzuki E, Sato Y. Hierarchical Gaussian Descriptor for Person Re-identification. In Proceedings - 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016. IEEE Computer Society. 2016. p. 1363-1372. 7780521. (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition). https://doi.org/10.1109/CVPR.2016.152
Matsukawa, Tetsu ; Okabe, Takahiro ; Suzuki, Einoshin ; Sato, Yoichi. / Hierarchical Gaussian Descriptor for Person Re-identification. Proceedings - 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016. IEEE Computer Society, 2016. pp. 1363-1372 (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition).
@inproceedings{af86d2716973499bb1176ea896fe0df4,
title = "Hierarchical Gaussian Descriptor for Person Re-identification",
abstract = "Describing the color and textural information of a person image is one of the most crucial aspects of person re-identification. In this paper, we present a novel descriptor based on a hierarchical distribution of pixel features. A hierarchical covariance descriptor has been successfully applied for image classification. However, the mean information of pixel features, which is absent in covariance, tends to be major discriminative information of person images. To solve this problem, we describe a local region in an image via hierarchical Gaussian distribution in which both means and covariances are included in their parameters. More specifically, we model the region as a set of multiple Gaussian distributions in which each Gaussian represents the appearance of a local patch. The characteristics of the set of Gaussians are again described by another Gaussian distribution. In both steps, unlike the hierarchical covariance descriptor, the proposed descriptor can model both the mean and the covariance information of pixel features properly. The results of experiments conducted on five databases indicate that the proposed descriptor exhibits re-markably high performance which outperforms the state-of-the-art descriptors for person re-identification.",
author = "Tetsu Matsukawa and Takahiro Okabe and Einoshin Suzuki and Yoichi Sato",
year = "2016",
month = "12",
day = "9",
doi = "10.1109/CVPR.2016.152",
language = "English",
series = "Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition",
publisher = "IEEE Computer Society",
pages = "1363--1372",
booktitle = "Proceedings - 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016",
address = "United States",

}

TY - GEN

T1 - Hierarchical Gaussian Descriptor for Person Re-identification

AU - Matsukawa, Tetsu

AU - Okabe, Takahiro

AU - Suzuki, Einoshin

AU - Sato, Yoichi

PY - 2016/12/9

Y1 - 2016/12/9

N2 - Describing the color and textural information of a person image is one of the most crucial aspects of person re-identification. In this paper, we present a novel descriptor based on a hierarchical distribution of pixel features. A hierarchical covariance descriptor has been successfully applied for image classification. However, the mean information of pixel features, which is absent in covariance, tends to be major discriminative information of person images. To solve this problem, we describe a local region in an image via hierarchical Gaussian distribution in which both means and covariances are included in their parameters. More specifically, we model the region as a set of multiple Gaussian distributions in which each Gaussian represents the appearance of a local patch. The characteristics of the set of Gaussians are again described by another Gaussian distribution. In both steps, unlike the hierarchical covariance descriptor, the proposed descriptor can model both the mean and the covariance information of pixel features properly. The results of experiments conducted on five databases indicate that the proposed descriptor exhibits re-markably high performance which outperforms the state-of-the-art descriptors for person re-identification.

AB - Describing the color and textural information of a person image is one of the most crucial aspects of person re-identification. In this paper, we present a novel descriptor based on a hierarchical distribution of pixel features. A hierarchical covariance descriptor has been successfully applied for image classification. However, the mean information of pixel features, which is absent in covariance, tends to be major discriminative information of person images. To solve this problem, we describe a local region in an image via hierarchical Gaussian distribution in which both means and covariances are included in their parameters. More specifically, we model the region as a set of multiple Gaussian distributions in which each Gaussian represents the appearance of a local patch. The characteristics of the set of Gaussians are again described by another Gaussian distribution. In both steps, unlike the hierarchical covariance descriptor, the proposed descriptor can model both the mean and the covariance information of pixel features properly. The results of experiments conducted on five databases indicate that the proposed descriptor exhibits re-markably high performance which outperforms the state-of-the-art descriptors for person re-identification.

UR - http://www.scopus.com/inward/record.url?scp=84986331442&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84986331442&partnerID=8YFLogxK

U2 - 10.1109/CVPR.2016.152

DO - 10.1109/CVPR.2016.152

M3 - Conference contribution

AN - SCOPUS:84986331442

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

SP - 1363

EP - 1372

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

PB - IEEE Computer Society

ER -