Real-Time Human Detection Using Hierarchical HOG Matrices

Guan PANG  Guijin WANG  Xinggang LIN  

IEICE TRANSACTIONS on Information and Systems   Vol.E93-D   No.3   pp.658-661
Publication Date: 2010/03/01
Online ISSN: 1745-1361
DOI: 10.1587/transinf.E93.D.658
Print ISSN: 0916-8532
Type of Manuscript: LETTER
Category: Image Recognition, Computer Vision
human detection,  real-time detection,  HOG,  window-scanning,  multi-detector,  

Full Text: PDF(485KB)>>
Buy this Article

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.