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.
Optimized Implementation of Pedestrian Tracking Using Multiple Cues on GPU
Ryusuke MIYAMOTO Hiroki SUGANO
IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences
Publication Date: 2011/11/01
Online ISSN: 1745-1337
Print ISSN: 0916-8508
Type of Manuscript: Special Section PAPER (Special Section on Smart Multimedia & Communication Systems)
Category: Image Processing
pedestrian tracking, GPU implementation, parallel processing, particle filter,
Full Text: PDF(14MB)>>
Nowadays, pedestrian recognition for automotive and security applications that require accurate recognition in images taken from distant observation points is a recent challenging problem in the field of computer vision. To achieve accurate recognition, both detection and tracking must be precise. For detection, some excellent schemes suitable for pedestrian recognition from distant observation points are proposed, however, no tracking schemes can achieve sufficient performance. To construct an accurate tracking scheme suitable for pedestrian recognition from distant observation points, we propose a novel pedestrian tracking scheme using multiple cues: HSV histograms and HOG features. Experimental results show that the proposed scheme can properly track a target pedestrian where tracking schemes using only a single cue fails. Moreover, we implement the proposed scheme on NVIDIA® TeslaTM C1060 processor, one of the latest GPU, to achieve real-time processing of the proposed scheme. Experimental results show that computation time required for tracking of a frame by our implementation is reduced to 8.80 ms even though Intel® CoreTM i7 CPU 975 @ 3.33 GHz spends 111 ms.