Real-Time Human Detection Using Hierarchical HOG Matrices

Guan PANG; Guijin WANG; Xinggang LIN

doi:10.1587/transinf.E93.D.658

Real-Time Human Detection Using Hierarchical HOG Matrices

Guan PANG, Guijin WANG, Xinggang LIN

Full Text Views

0

Cite this

Summary :

Human detection has witnessed significant development in recent years. The introduction of cascade structure and integral histogram has greatly improved detection speed. But real-time detection is still only possible for sparse scan of 320 240 sized images. In this work, we propose a matrix-based structure to reorganize the computation structure of window-scanning detection algorithms, as well as a new pre-processing method called Hierarchical HOG Matrices (HHM) in place of integral histogram. Our speed-up scheme can process 320 240 sized images by dense scan (≈ 12000 windows per image) at the speed of about 30 fps, while maintaining accuracy comparable to the original HOG + cascade method.

Publication: IEICE TRANSACTIONS on Information Vol.E93-D No.3 pp.658-661

Publication Date: 2010/03/01

Publicized

Online ISSN: 1745-1361

DOI: 10.1587/transinf.E93.D.658

Type of Manuscript: LETTER

Category: Image Recognition, Computer Vision

Cite this

Copy

Guan PANG, Guijin WANG, Xinggang LIN, "Real-Time Human Detection Using Hierarchical HOG Matrices" in IEICE TRANSACTIONS on Information, vol. E93-D, no. 3, pp. 658-661, March 2010, doi: 10.1587/transinf.E93.D.658.
Abstract: Human detection has witnessed significant development in recent years. The introduction of cascade structure and integral histogram has greatly improved detection speed. But real-time detection is still only possible for sparse scan of 320 240 sized images. In this work, we propose a matrix-based structure to reorganize the computation structure of window-scanning detection algorithms, as well as a new pre-processing method called Hierarchical HOG Matrices (HHM) in place of integral histogram. Our speed-up scheme can process 320 240 sized images by dense scan (≈ 12000 windows per image) at the speed of about 30 fps, while maintaining accuracy comparable to the original HOG + cascade method.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E93.D.658/_p

Copy

@ARTICLE{e93-d_3_658,
author={Guan PANG, Guijin WANG, Xinggang LIN, },
journal={IEICE TRANSACTIONS on Information},
title={Real-Time Human Detection Using Hierarchical HOG Matrices},
year={2010},
volume={E93-D},
number={3},
pages={658-661},
abstract={Human detection has witnessed significant development in recent years. The introduction of cascade structure and integral histogram has greatly improved detection speed. But real-time detection is still only possible for sparse scan of 320 240 sized images. In this work, we propose a matrix-based structure to reorganize the computation structure of window-scanning detection algorithms, as well as a new pre-processing method called Hierarchical HOG Matrices (HHM) in place of integral histogram. Our speed-up scheme can process 320 240 sized images by dense scan (≈ 12000 windows per image) at the speed of about 30 fps, while maintaining accuracy comparable to the original HOG + cascade method.},
keywords={},
doi={10.1587/transinf.E93.D.658},
ISSN={1745-1361},
month={March},}

Copy

TY - JOUR
TI - Real-Time Human Detection Using Hierarchical HOG Matrices
T2 - IEICE TRANSACTIONS on Information
SP - 658
EP - 661
AU - Guan PANG
AU - Guijin WANG
AU - Xinggang LIN
PY - 2010
DO - 10.1587/transinf.E93.D.658
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E93-D
IS - 3
JA - IEICE TRANSACTIONS on Information
Y1 - March 2010
AB - Human detection has witnessed significant development in recent years. The introduction of cascade structure and integral histogram has greatly improved detection speed. But real-time detection is still only possible for sparse scan of 320 240 sized images. In this work, we propose a matrix-based structure to reorganize the computation structure of window-scanning detection algorithms, as well as a new pre-processing method called Hierarchical HOG Matrices (HHM) in place of integral histogram. Our speed-up scheme can process 320 240 sized images by dense scan (≈ 12000 windows per image) at the speed of about 30 fps, while maintaining accuracy comparable to the original HOG + cascade method.
ER -