Cargando…

Integral Histogram with Random Projection for Pedestrian Detection

In this paper, we give a systematic study to report several deep insights into the HOG, one of the most widely used features in the modern computer vision and image processing applications. We first show that, its magnitudes of gradient can be randomly projected with random matrix. To handle over-fi...

Descripción completa

Detalles Bibliográficos
Autores principales: Liu, Chang-Hua, Lin, Jian-Kun
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Public Library of Science 2015
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4646677/
https://www.ncbi.nlm.nih.gov/pubmed/26569486
http://dx.doi.org/10.1371/journal.pone.0142820
_version_ 1782400972421595136
author Liu, Chang-Hua
Lin, Jian-Kun
author_facet Liu, Chang-Hua
Lin, Jian-Kun
author_sort Liu, Chang-Hua
collection PubMed
description In this paper, we give a systematic study to report several deep insights into the HOG, one of the most widely used features in the modern computer vision and image processing applications. We first show that, its magnitudes of gradient can be randomly projected with random matrix. To handle over-fitting, an integral histogram based on the differences of randomly selected blocks is proposed. The experiments show that both the random projection and integral histogram outperform the HOG feature obviously. Finally, the two ideas are combined into a new descriptor termed IHRP, which outperforms the HOG feature with less dimensions and higher speed.
format Online
Article
Text
id pubmed-4646677
institution National Center for Biotechnology Information
language English
publishDate 2015
publisher Public Library of Science
record_format MEDLINE/PubMed
spelling pubmed-46466772015-11-25 Integral Histogram with Random Projection for Pedestrian Detection Liu, Chang-Hua Lin, Jian-Kun PLoS One Research Article In this paper, we give a systematic study to report several deep insights into the HOG, one of the most widely used features in the modern computer vision and image processing applications. We first show that, its magnitudes of gradient can be randomly projected with random matrix. To handle over-fitting, an integral histogram based on the differences of randomly selected blocks is proposed. The experiments show that both the random projection and integral histogram outperform the HOG feature obviously. Finally, the two ideas are combined into a new descriptor termed IHRP, which outperforms the HOG feature with less dimensions and higher speed. Public Library of Science 2015-11-16 /pmc/articles/PMC4646677/ /pubmed/26569486 http://dx.doi.org/10.1371/journal.pone.0142820 Text en © 2015 Liu, Lin http://creativecommons.org/licenses/by/4.0/ This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
spellingShingle Research Article
Liu, Chang-Hua
Lin, Jian-Kun
Integral Histogram with Random Projection for Pedestrian Detection
title Integral Histogram with Random Projection for Pedestrian Detection
title_full Integral Histogram with Random Projection for Pedestrian Detection
title_fullStr Integral Histogram with Random Projection for Pedestrian Detection
title_full_unstemmed Integral Histogram with Random Projection for Pedestrian Detection
title_short Integral Histogram with Random Projection for Pedestrian Detection
title_sort integral histogram with random projection for pedestrian detection
topic Research Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4646677/
https://www.ncbi.nlm.nih.gov/pubmed/26569486
http://dx.doi.org/10.1371/journal.pone.0142820
work_keys_str_mv AT liuchanghua integralhistogramwithrandomprojectionforpedestriandetection
AT linjiankun integralhistogramwithrandomprojectionforpedestriandetection