Histogram of template for pedestrian detection

Shaopeng Tang, Satoshi Goto

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

In this paper, we propose a novel feature named histogram of template (HOT) for human detection in still images. For every pixel of an image, various templates are defined, each of which contains the pixel itself and two of its neighboring pixels. If the texture and gradient values of the three pixels satisfy a pre-defined formula, the central pixel is regarded to meet the corresponding template for this formula. Histograms of pixels meeting various templates are calculated for a set of formulas, and combined to be the feature for detection. Compared to the other features, the proposed feature takes texture as well as the gradient information into consideration. Besides, it reflects the relationship between 3 pixels, instead of focusing on only one. Experiments for human detection are performed on INRIA dataset, which shows the proposed HOT feature is more discriminative than histogram of orientated gradient (HOG) feature, under the same training method.

Original languageEnglish
Pages (from-to)1737-1744
Number of pages8
JournalIEICE Transactions on Information and Systems
VolumeE93-D
Issue number7
DOIs
Publication statusPublished - 2010 Jul

Fingerprint

Pixels
Textures
Experiments

Keywords

  • Histogram of template
  • Human detection
  • SVM

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Software
  • Artificial Intelligence
  • Hardware and Architecture
  • Computer Vision and Pattern Recognition

Cite this

Histogram of template for pedestrian detection. / Tang, Shaopeng; Goto, Satoshi.

In: IEICE Transactions on Information and Systems, Vol. E93-D, No. 7, 07.2010, p. 1737-1744.

Research output: Contribution to journalArticle

Tang, Shaopeng ; Goto, Satoshi. / Histogram of template for pedestrian detection. In: IEICE Transactions on Information and Systems. 2010 ; Vol. E93-D, No. 7. pp. 1737-1744.
@article{34d60861ca74439ea0905692b8f4d23e,
title = "Histogram of template for pedestrian detection",
abstract = "In this paper, we propose a novel feature named histogram of template (HOT) for human detection in still images. For every pixel of an image, various templates are defined, each of which contains the pixel itself and two of its neighboring pixels. If the texture and gradient values of the three pixels satisfy a pre-defined formula, the central pixel is regarded to meet the corresponding template for this formula. Histograms of pixels meeting various templates are calculated for a set of formulas, and combined to be the feature for detection. Compared to the other features, the proposed feature takes texture as well as the gradient information into consideration. Besides, it reflects the relationship between 3 pixels, instead of focusing on only one. Experiments for human detection are performed on INRIA dataset, which shows the proposed HOT feature is more discriminative than histogram of orientated gradient (HOG) feature, under the same training method.",
keywords = "Histogram of template, Human detection, SVM",
author = "Shaopeng Tang and Satoshi Goto",
year = "2010",
month = "7",
doi = "10.1587/transinf.E93.D.1737",
language = "English",
volume = "E93-D",
pages = "1737--1744",
journal = "IEICE Transactions on Information and Systems",
issn = "0916-8532",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
number = "7",

}

TY - JOUR

T1 - Histogram of template for pedestrian detection

AU - Tang, Shaopeng

AU - Goto, Satoshi

PY - 2010/7

Y1 - 2010/7

N2 - In this paper, we propose a novel feature named histogram of template (HOT) for human detection in still images. For every pixel of an image, various templates are defined, each of which contains the pixel itself and two of its neighboring pixels. If the texture and gradient values of the three pixels satisfy a pre-defined formula, the central pixel is regarded to meet the corresponding template for this formula. Histograms of pixels meeting various templates are calculated for a set of formulas, and combined to be the feature for detection. Compared to the other features, the proposed feature takes texture as well as the gradient information into consideration. Besides, it reflects the relationship between 3 pixels, instead of focusing on only one. Experiments for human detection are performed on INRIA dataset, which shows the proposed HOT feature is more discriminative than histogram of orientated gradient (HOG) feature, under the same training method.

AB - In this paper, we propose a novel feature named histogram of template (HOT) for human detection in still images. For every pixel of an image, various templates are defined, each of which contains the pixel itself and two of its neighboring pixels. If the texture and gradient values of the three pixels satisfy a pre-defined formula, the central pixel is regarded to meet the corresponding template for this formula. Histograms of pixels meeting various templates are calculated for a set of formulas, and combined to be the feature for detection. Compared to the other features, the proposed feature takes texture as well as the gradient information into consideration. Besides, it reflects the relationship between 3 pixels, instead of focusing on only one. Experiments for human detection are performed on INRIA dataset, which shows the proposed HOT feature is more discriminative than histogram of orientated gradient (HOG) feature, under the same training method.

KW - Histogram of template

KW - Human detection

KW - SVM

UR - http://www.scopus.com/inward/record.url?scp=77956027632&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77956027632&partnerID=8YFLogxK

U2 - 10.1587/transinf.E93.D.1737

DO - 10.1587/transinf.E93.D.1737

M3 - Article

VL - E93-D

SP - 1737

EP - 1744

JO - IEICE Transactions on Information and Systems

JF - IEICE Transactions on Information and Systems

SN - 0916-8532

IS - 7

ER -