For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Efficient Pedestrian Detection Using Multi-Scale HOG Features with Low Computational Complexity
Soojin KIM Kyeongsoon CHO
IEICE TRANSACTIONS on Information and Systems
Publication Date: 2014/02/01
Online ISSN: 1745-1361
Print ISSN: 0916-8532
Type of Manuscript: LETTER
Category: Pattern Recognition
pedestrian detection, interpolation, multi-scale HOG, integral HOG, two-stage approach,
Full Text: PDF(577.2KB)>>
In this paper, an efficient method to reduce computational complexity for pedestrian detection is presented. Since trilinear interpolation is not used, the amount of required operations for histogram of oriented gradient (HOG) feature calculation is significantly reduced. By calculating multi-scale HOG features with integral HOG in a two-stage approach, both high detection rate and speed are achieved in the proposed method.