Evrişimsel Sinir Aǧi Öznitelikleri ile Kişiyi Yeniden Tanima

Alper Ulu; Hazim Kemal Ekenel

doi:10.1109/SIU.2016.7495897

Evrişimsel Sinir Aǧi Öznitelikleri ile Kişiyi Yeniden Tanima

Translated title of the contribution: Convolutional neural network-based representation for person re-identification

Alper Ulu, Hazim Kemal Ekenel

Department of Computer Engineering

Istanbul Technical University

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

5 Citations (Scopus)

Abstract

In this paper, using a general convolutional neural network (CNN) model, which was developed for object recognition, a successful system has been introduced for the person re-identification problem. To use this CNN model for the person re-identification problem properly, it is individually fine-tuned using different body parts of person images. For feature extraction, we used the seventh layer of the CNN model, which was re-trained with the available datasets. Then, we used cosine similarity metric to calculate the similarity between extracted features. CUHK03 and Market-1501 datasets were used as the training sets and the proposed method has been tested on VIPeR dataset. Superior results have been obtained with the proposed method, compared to the state-of-the-art methods in the field.

Translated title of the contribution	Convolutional neural network-based representation for person re-identification
Original language	Turkish
Title of host publication	2016 24th Signal Processing and Communication Application Conference, SIU 2016 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	945-948
Number of pages	4
ISBN (Electronic)	9781509016792
DOIs	https://doi.org/10.1109/SIU.2016.7495897
Publication status	Published - 20 Jun 2016
Event	24th Signal Processing and Communication Application Conference, SIU 2016 - Zonguldak, Turkey Duration: 16 May 2016 → 19 May 2016

Publication series

Name	2016 24th Signal Processing and Communication Application Conference, SIU 2016 - Proceedings

Conference

Conference	24th Signal Processing and Communication Application Conference, SIU 2016
Country/Territory	Turkey
City	Zonguldak
Period	16/05/16 → 19/05/16

Bibliographical note

Publisher Copyright:
© 2016 IEEE.

Access to Document

10.1109/SIU.2016.7495897

Cite this

Ulu, A., & Ekenel, H. K. (2016). Evrişimsel Sinir Aǧi Öznitelikleri ile Kişiyi Yeniden Tanima. In 2016 24th Signal Processing and Communication Application Conference, SIU 2016 - Proceedings (pp. 945-948). Article 7495897 (2016 24th Signal Processing and Communication Application Conference, SIU 2016 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SIU.2016.7495897

@inproceedings{612f3c2dfb824cce8636b6d7b01132ed,

title = "Evri{\c s}imsel Sinir Aǧi {\"O}znitelikleri ile Ki{\c s}iyi Yeniden Tanima",

abstract = "In this paper, using a general convolutional neural network (CNN) model, which was developed for object recognition, a successful system has been introduced for the person re-identification problem. To use this CNN model for the person re-identification problem properly, it is individually fine-tuned using different body parts of person images. For feature extraction, we used the seventh layer of the CNN model, which was re-trained with the available datasets. Then, we used cosine similarity metric to calculate the similarity between extracted features. CUHK03 and Market-1501 datasets were used as the training sets and the proposed method has been tested on VIPeR dataset. Superior results have been obtained with the proposed method, compared to the state-of-the-art methods in the field.",

keywords = "convolutional neural network, cosine similarity, person re-identification",

author = "Alper Ulu and Ekenel, {Hazim Kemal}",

note = "Publisher Copyright: {\textcopyright} 2016 IEEE.; 24th Signal Processing and Communication Application Conference, SIU 2016 ; Conference date: 16-05-2016 Through 19-05-2016",

year = "2016",

month = jun,

day = "20",

doi = "10.1109/SIU.2016.7495897",

language = "T{\"u}rk{\c c}e",

series = "2016 24th Signal Processing and Communication Application Conference, SIU 2016 - Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "945--948",

booktitle = "2016 24th Signal Processing and Communication Application Conference, SIU 2016 - Proceedings",

address = "United States",

}

Ulu, A & Ekenel, HK 2016, Evrişimsel Sinir Aǧi Öznitelikleri ile Kişiyi Yeniden Tanima. in 2016 24th Signal Processing and Communication Application Conference, SIU 2016 - Proceedings., 7495897, 2016 24th Signal Processing and Communication Application Conference, SIU 2016 - Proceedings, Institute of Electrical and Electronics Engineers Inc., pp. 945-948, 24th Signal Processing and Communication Application Conference, SIU 2016, Zonguldak, Turkey, 16/05/16. https://doi.org/10.1109/SIU.2016.7495897

Evrişimsel Sinir Aǧi Öznitelikleri ile Kişiyi Yeniden Tanima. / Ulu, Alper; Ekenel, Hazim Kemal.
2016 24th Signal Processing and Communication Application Conference, SIU 2016 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2016. p. 945-948 7495897 (2016 24th Signal Processing and Communication Application Conference, SIU 2016 - Proceedings).

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

TY - GEN

T1 - Evrişimsel Sinir Aǧi Öznitelikleri ile Kişiyi Yeniden Tanima

AU - Ulu, Alper

AU - Ekenel, Hazim Kemal

PY - 2016/6/20

Y1 - 2016/6/20

N2 - In this paper, using a general convolutional neural network (CNN) model, which was developed for object recognition, a successful system has been introduced for the person re-identification problem. To use this CNN model for the person re-identification problem properly, it is individually fine-tuned using different body parts of person images. For feature extraction, we used the seventh layer of the CNN model, which was re-trained with the available datasets. Then, we used cosine similarity metric to calculate the similarity between extracted features. CUHK03 and Market-1501 datasets were used as the training sets and the proposed method has been tested on VIPeR dataset. Superior results have been obtained with the proposed method, compared to the state-of-the-art methods in the field.

AB - In this paper, using a general convolutional neural network (CNN) model, which was developed for object recognition, a successful system has been introduced for the person re-identification problem. To use this CNN model for the person re-identification problem properly, it is individually fine-tuned using different body parts of person images. For feature extraction, we used the seventh layer of the CNN model, which was re-trained with the available datasets. Then, we used cosine similarity metric to calculate the similarity between extracted features. CUHK03 and Market-1501 datasets were used as the training sets and the proposed method has been tested on VIPeR dataset. Superior results have been obtained with the proposed method, compared to the state-of-the-art methods in the field.

KW - convolutional neural network

KW - cosine similarity

KW - person re-identification

UR - http://www.scopus.com/inward/record.url?scp=84982806149&partnerID=8YFLogxK

U2 - 10.1109/SIU.2016.7495897

DO - 10.1109/SIU.2016.7495897

M3 - Konferans katkısı

AN - SCOPUS:84982806149

T3 - 2016 24th Signal Processing and Communication Application Conference, SIU 2016 - Proceedings

SP - 945

EP - 948

BT - 2016 24th Signal Processing and Communication Application Conference, SIU 2016 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 24th Signal Processing and Communication Application Conference, SIU 2016

Y2 - 16 May 2016 through 19 May 2016

ER -

Ulu A, Ekenel HK. Evrişimsel Sinir Aǧi Öznitelikleri ile Kişiyi Yeniden Tanima. In 2016 24th Signal Processing and Communication Application Conference, SIU 2016 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2016. p. 945-948. 7495897. (2016 24th Signal Processing and Communication Application Conference, SIU 2016 - Proceedings). doi: 10.1109/SIU.2016.7495897

Evrişimsel Sinir Aǧi Öznitelikleri ile Kişiyi Yeniden Tanima

Abstract

Publication series

Conference

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this