In this paper, a nonoverlapping multi-camera and people re-identification algorithm is proposed. It applies inflated major color features for re-identification to reduce computation time. The inflated major color features can dramatically improve efficiency while retaining high accuracy of object re-identification. The proposed method is evaluated over a wide range of experimental databases. The accuracy attains upwards of 40.7% in Rank 1 and 84% in Rank 10 on average, while it obtains three to 15 times faster than algorithms reported in the literature. The proposed algorithm has been implemented on a SOC-FPGA platform to reach 50 FPS with 1280×720 HD resolution and 25 FPS with 1920×1080 FHD resolution for real-time processing. The results show a performance improvement and reduction in computation complexity, which is especially ideal for embedded platform.
Chi-Chia SUN
Department of Electrical Engineering National Formosa University
Ming-Hwa SHEU
Department of Electronic Engineering National Yunlin University of Science and Technology
Jui-Yang CHI
Department of Electronic Engineering National Yunlin University of Science and Technology
Yan-Kai HUANG
Department of Electronic Engineering National Yunlin University of Science and Technology
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Chi-Chia SUN, Ming-Hwa SHEU, Jui-Yang CHI, Yan-Kai HUANG, "A Fast Non-Overlapping Multi-Camera People Re-Identification Algorithm and Tracking Based on Visual Channel Model" in IEICE TRANSACTIONS on Information,
vol. E102-D, no. 7, pp. 1342-1348, July 2019, doi: 10.1587/transinf.2018EDP7348.
Abstract: In this paper, a nonoverlapping multi-camera and people re-identification algorithm is proposed. It applies inflated major color features for re-identification to reduce computation time. The inflated major color features can dramatically improve efficiency while retaining high accuracy of object re-identification. The proposed method is evaluated over a wide range of experimental databases. The accuracy attains upwards of 40.7% in Rank 1 and 84% in Rank 10 on average, while it obtains three to 15 times faster than algorithms reported in the literature. The proposed algorithm has been implemented on a SOC-FPGA platform to reach 50 FPS with 1280×720 HD resolution and 25 FPS with 1920×1080 FHD resolution for real-time processing. The results show a performance improvement and reduction in computation complexity, which is especially ideal for embedded platform.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2018EDP7348/_p
Copy
@ARTICLE{e102-d_7_1342,
author={Chi-Chia SUN, Ming-Hwa SHEU, Jui-Yang CHI, Yan-Kai HUANG, },
journal={IEICE TRANSACTIONS on Information},
title={A Fast Non-Overlapping Multi-Camera People Re-Identification Algorithm and Tracking Based on Visual Channel Model},
year={2019},
volume={E102-D},
number={7},
pages={1342-1348},
abstract={In this paper, a nonoverlapping multi-camera and people re-identification algorithm is proposed. It applies inflated major color features for re-identification to reduce computation time. The inflated major color features can dramatically improve efficiency while retaining high accuracy of object re-identification. The proposed method is evaluated over a wide range of experimental databases. The accuracy attains upwards of 40.7% in Rank 1 and 84% in Rank 10 on average, while it obtains three to 15 times faster than algorithms reported in the literature. The proposed algorithm has been implemented on a SOC-FPGA platform to reach 50 FPS with 1280×720 HD resolution and 25 FPS with 1920×1080 FHD resolution for real-time processing. The results show a performance improvement and reduction in computation complexity, which is especially ideal for embedded platform.},
keywords={},
doi={10.1587/transinf.2018EDP7348},
ISSN={1745-1361},
month={July},}
Copy
TY - JOUR
TI - A Fast Non-Overlapping Multi-Camera People Re-Identification Algorithm and Tracking Based on Visual Channel Model
T2 - IEICE TRANSACTIONS on Information
SP - 1342
EP - 1348
AU - Chi-Chia SUN
AU - Ming-Hwa SHEU
AU - Jui-Yang CHI
AU - Yan-Kai HUANG
PY - 2019
DO - 10.1587/transinf.2018EDP7348
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E102-D
IS - 7
JA - IEICE TRANSACTIONS on Information
Y1 - July 2019
AB - In this paper, a nonoverlapping multi-camera and people re-identification algorithm is proposed. It applies inflated major color features for re-identification to reduce computation time. The inflated major color features can dramatically improve efficiency while retaining high accuracy of object re-identification. The proposed method is evaluated over a wide range of experimental databases. The accuracy attains upwards of 40.7% in Rank 1 and 84% in Rank 10 on average, while it obtains three to 15 times faster than algorithms reported in the literature. The proposed algorithm has been implemented on a SOC-FPGA platform to reach 50 FPS with 1280×720 HD resolution and 25 FPS with 1920×1080 FHD resolution for real-time processing. The results show a performance improvement and reduction in computation complexity, which is especially ideal for embedded platform.
ER -