Bayesian learning of a search region for pedestrian detection

Jeonghyun Baek, Sungjun Hong, Jisu Kim, Euntai Kim

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

An efficient pedestrian detection method is proposed for intelligent vehicles in this paper. The proposed method learns the region in which pedestrians are likely to be detected and narrows down the search to the likely region. The likely region is modeled as a Gaussian distribution on the y-axis and its parameters are updated by a Bayesian approach. Thus, the proposed method starts with an exhaustive full search, but gradually narrows down the search by focusing on the likely region. The learning of the likely region is formulated as a Bayesian learning problem and the likely region is analytically derived. The proposed method is combined with two popular pedestrian detection methods, Haar-like Adaboost and HOG-LSVM, and some experiments are conducted with the Caltech pedestrian dataset. The experiments show that the proposed method not only reduces computation time, but also enhances performance by rejecting false positive results.

Original languageEnglish
Pages (from-to)863-885
Number of pages23
JournalMultimedia Tools and Applications
Volume75
Issue number2
DOIs
Publication statusPublished - 2016 Jan 1

Fingerprint

Intelligent vehicle highway systems
Adaptive boosting
Gaussian distribution
Experiments

All Science Journal Classification (ASJC) codes

  • Software
  • Media Technology
  • Hardware and Architecture
  • Computer Networks and Communications

Cite this

Baek, Jeonghyun ; Hong, Sungjun ; Kim, Jisu ; Kim, Euntai. / Bayesian learning of a search region for pedestrian detection. In: Multimedia Tools and Applications. 2016 ; Vol. 75, No. 2. pp. 863-885.
@article{e7087bfbd73d400ab8b059d241c71c07,
title = "Bayesian learning of a search region for pedestrian detection",
abstract = "An efficient pedestrian detection method is proposed for intelligent vehicles in this paper. The proposed method learns the region in which pedestrians are likely to be detected and narrows down the search to the likely region. The likely region is modeled as a Gaussian distribution on the y-axis and its parameters are updated by a Bayesian approach. Thus, the proposed method starts with an exhaustive full search, but gradually narrows down the search by focusing on the likely region. The learning of the likely region is formulated as a Bayesian learning problem and the likely region is analytically derived. The proposed method is combined with two popular pedestrian detection methods, Haar-like Adaboost and HOG-LSVM, and some experiments are conducted with the Caltech pedestrian dataset. The experiments show that the proposed method not only reduces computation time, but also enhances performance by rejecting false positive results.",
author = "Jeonghyun Baek and Sungjun Hong and Jisu Kim and Euntai Kim",
year = "2016",
month = "1",
day = "1",
doi = "10.1007/s11042-014-2329-z",
language = "English",
volume = "75",
pages = "863--885",
journal = "Multimedia Tools and Applications",
issn = "1380-7501",
publisher = "Springer Netherlands",
number = "2",

}

Bayesian learning of a search region for pedestrian detection. / Baek, Jeonghyun; Hong, Sungjun; Kim, Jisu; Kim, Euntai.

In: Multimedia Tools and Applications, Vol. 75, No. 2, 01.01.2016, p. 863-885.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Bayesian learning of a search region for pedestrian detection

AU - Baek, Jeonghyun

AU - Hong, Sungjun

AU - Kim, Jisu

AU - Kim, Euntai

PY - 2016/1/1

Y1 - 2016/1/1

N2 - An efficient pedestrian detection method is proposed for intelligent vehicles in this paper. The proposed method learns the region in which pedestrians are likely to be detected and narrows down the search to the likely region. The likely region is modeled as a Gaussian distribution on the y-axis and its parameters are updated by a Bayesian approach. Thus, the proposed method starts with an exhaustive full search, but gradually narrows down the search by focusing on the likely region. The learning of the likely region is formulated as a Bayesian learning problem and the likely region is analytically derived. The proposed method is combined with two popular pedestrian detection methods, Haar-like Adaboost and HOG-LSVM, and some experiments are conducted with the Caltech pedestrian dataset. The experiments show that the proposed method not only reduces computation time, but also enhances performance by rejecting false positive results.

AB - An efficient pedestrian detection method is proposed for intelligent vehicles in this paper. The proposed method learns the region in which pedestrians are likely to be detected and narrows down the search to the likely region. The likely region is modeled as a Gaussian distribution on the y-axis and its parameters are updated by a Bayesian approach. Thus, the proposed method starts with an exhaustive full search, but gradually narrows down the search by focusing on the likely region. The learning of the likely region is formulated as a Bayesian learning problem and the likely region is analytically derived. The proposed method is combined with two popular pedestrian detection methods, Haar-like Adaboost and HOG-LSVM, and some experiments are conducted with the Caltech pedestrian dataset. The experiments show that the proposed method not only reduces computation time, but also enhances performance by rejecting false positive results.

UR - http://www.scopus.com/inward/record.url?scp=84955754596&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84955754596&partnerID=8YFLogxK

U2 - 10.1007/s11042-014-2329-z

DO - 10.1007/s11042-014-2329-z

M3 - Article

AN - SCOPUS:84955754596

VL - 75

SP - 863

EP - 885

JO - Multimedia Tools and Applications

JF - Multimedia Tools and Applications

SN - 1380-7501

IS - 2

ER -