Pedestrian detection based on bidirectional local template patterns

Jiu Xu, Ning Jiang, Satoshi Goto

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

3 Citations (Scopus)

Abstract

In this paper, a novel feature named bidirectional local template patterns (B-LTP) is proposed to achieve pedestrian detection in still image. This feature is a combined and modified version of histogram of template (HOT) [1] and center-symmetric local binary patterns (CS-LBP) [2]. For each pixel, four templates are defined, each of which contains the pixel itself and two of its neighboring center-symmetric pixels. For each template, not only the relationships between three pixels according to the template, but also information of two directions are calculated in our feature, which makes it more discriminative. Moreover, the feature length of B-LTP is very short, which costs less computational workload and memory consumption. Experimental results on INRIA dataset show that both the speed and detection rate of our proposed B-LTP feature outperform other features such as histogram of orientated gradient (HOG) [3], HOT and Covariance Matrix (COV) [4].

Original languageEnglish
Title of host publicationEuropean Signal Processing Conference
Pages400-404
Number of pages5
Publication statusPublished - 2012
Event20th European Signal Processing Conference, EUSIPCO 2012 - Bucharest
Duration: 2012 Aug 272012 Aug 31

Other

Other20th European Signal Processing Conference, EUSIPCO 2012
CityBucharest
Period12/8/2712/8/31

Fingerprint

Pixels
Covariance matrix
Data storage equipment
Costs

Keywords

  • bidirectional local template patterns
  • Pedestrian detection
  • support vector machine

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Xu, J., Jiang, N., & Goto, S. (2012). Pedestrian detection based on bidirectional local template patterns. In European Signal Processing Conference (pp. 400-404). [6333928]

Pedestrian detection based on bidirectional local template patterns. / Xu, Jiu; Jiang, Ning; Goto, Satoshi.

European Signal Processing Conference. 2012. p. 400-404 6333928.

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

Xu, J, Jiang, N & Goto, S 2012, Pedestrian detection based on bidirectional local template patterns. in European Signal Processing Conference., 6333928, pp. 400-404, 20th European Signal Processing Conference, EUSIPCO 2012, Bucharest, 12/8/27.
Xu J, Jiang N, Goto S. Pedestrian detection based on bidirectional local template patterns. In European Signal Processing Conference. 2012. p. 400-404. 6333928
Xu, Jiu ; Jiang, Ning ; Goto, Satoshi. / Pedestrian detection based on bidirectional local template patterns. European Signal Processing Conference. 2012. pp. 400-404
@inproceedings{c209067138924fb39932c2843da7c20e,
title = "Pedestrian detection based on bidirectional local template patterns",
abstract = "In this paper, a novel feature named bidirectional local template patterns (B-LTP) is proposed to achieve pedestrian detection in still image. This feature is a combined and modified version of histogram of template (HOT) [1] and center-symmetric local binary patterns (CS-LBP) [2]. For each pixel, four templates are defined, each of which contains the pixel itself and two of its neighboring center-symmetric pixels. For each template, not only the relationships between three pixels according to the template, but also information of two directions are calculated in our feature, which makes it more discriminative. Moreover, the feature length of B-LTP is very short, which costs less computational workload and memory consumption. Experimental results on INRIA dataset show that both the speed and detection rate of our proposed B-LTP feature outperform other features such as histogram of orientated gradient (HOG) [3], HOT and Covariance Matrix (COV) [4].",
keywords = "bidirectional local template patterns, Pedestrian detection, support vector machine",
author = "Jiu Xu and Ning Jiang and Satoshi Goto",
year = "2012",
language = "English",
isbn = "9781467310680",
pages = "400--404",
booktitle = "European Signal Processing Conference",

}

TY - GEN

T1 - Pedestrian detection based on bidirectional local template patterns

AU - Xu, Jiu

AU - Jiang, Ning

AU - Goto, Satoshi

PY - 2012

Y1 - 2012

N2 - In this paper, a novel feature named bidirectional local template patterns (B-LTP) is proposed to achieve pedestrian detection in still image. This feature is a combined and modified version of histogram of template (HOT) [1] and center-symmetric local binary patterns (CS-LBP) [2]. For each pixel, four templates are defined, each of which contains the pixel itself and two of its neighboring center-symmetric pixels. For each template, not only the relationships between three pixels according to the template, but also information of two directions are calculated in our feature, which makes it more discriminative. Moreover, the feature length of B-LTP is very short, which costs less computational workload and memory consumption. Experimental results on INRIA dataset show that both the speed and detection rate of our proposed B-LTP feature outperform other features such as histogram of orientated gradient (HOG) [3], HOT and Covariance Matrix (COV) [4].

AB - In this paper, a novel feature named bidirectional local template patterns (B-LTP) is proposed to achieve pedestrian detection in still image. This feature is a combined and modified version of histogram of template (HOT) [1] and center-symmetric local binary patterns (CS-LBP) [2]. For each pixel, four templates are defined, each of which contains the pixel itself and two of its neighboring center-symmetric pixels. For each template, not only the relationships between three pixels according to the template, but also information of two directions are calculated in our feature, which makes it more discriminative. Moreover, the feature length of B-LTP is very short, which costs less computational workload and memory consumption. Experimental results on INRIA dataset show that both the speed and detection rate of our proposed B-LTP feature outperform other features such as histogram of orientated gradient (HOG) [3], HOT and Covariance Matrix (COV) [4].

KW - bidirectional local template patterns

KW - Pedestrian detection

KW - support vector machine

UR - http://www.scopus.com/inward/record.url?scp=84869784167&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84869784167&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9781467310680

SP - 400

EP - 404

BT - European Signal Processing Conference

ER -