Implementation of Scale and Rotation Invariant On-Line Object Tracking Based on CUDA

Quan MIAO  Guijin WANG  Xinggang LIN  

Publication
IEICE TRANSACTIONS on Information and Systems   Vol.E94-D   No.12   pp.2549-2552
Publication Date: 2011/12/01
Online ISSN: 1745-1361
DOI: 10.1587/transinf.E94.D.2549
Print ISSN: 0916-8532
Type of Manuscript: LETTER
Category: Image Recognition, Computer Vision
Keyword: 
object tracking,  classifier updating,  GPGPU,  CUDA,  

Full Text: PDF(791.4KB)>>
Buy this Article




Summary: 
Object tracking is a major technique in image processing and computer vision. Tracking speed will directly determine the quality of applications. This paper presents a parallel implementation for a recently proposed scale- and rotation-invariant on-line object tracking system. The algorithm is based on NVIDIA's Graphics Processing Units (GPU) using Compute Unified Device Architecture (CUDA), following the model of single instruction multiple threads. Specifically, we analyze the original algorithm and propose the GPU-based parallel design. Emphasis is placed on exploiting the data parallelism and memory usage. In addition, we apply optimization technique to maximize the utilization of NVIDIA's GPU and reduce the data transfer time. Experimental results show that our GPGPU-based method running on a GTX480 graphics card could achieve up to 12X speed-up compared with the efficiency equivalence on an Intel E8400 3.0 GHz CPU, including I/O time.