Visual tracking based on particle filter with spline resampling

被引:5
作者
Choe, Gwangmin [1 ]
Wang, Tianjiang [1 ]
Liu, Fang [1 ]
Choe, Chunhwa [2 ]
So, Hyoson [2 ]
机构
[1] Huazhong Univ Sci & Technol, Sch Comp Sci & Technol, Intelligent & Distributed Comp Lab, Wuhan 430074, Peoples R China
[2] Kim Il Sung Univ, Sch Comp Sci & Technol, Visual Informat Proc Lab, Pyongyang, North Korea
关键词
Spline resampling; Point spread function; Particle filter; Impoverishment; Visual tracking; ALGORITHM;
D O I
10.1007/s11042-014-1960-z
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
We introduce the concept of a spline resampling in the particle filter to deal with the high accuracy and the sample impoverishment. The resampling is usually based on a linear transformation on the weights of the particles, so it affects the accurate filtering. The spline resampling consists of two parts: the spline transformation of weights and the spread transformation of states. The former is based on a spline transformation on the weights of the particles to obtain the high accuracy of particle filtering, and the latter is based on a point spread transformation on states of particles to prevent the sample impoverishment due to decline of the diversity of hypothesis after resampling. Two transformations are sequentially implemented to incorporate with each other. Then, we propose a global transition model in the particle filter, which takes account of the background variation caused by the camera motion model of object itself, to decrease error from real object position. We test the performance of our spline resampling and the global transition model in the particle filter in object tracking scenario. Experimental results demonstrate that particle filter with the spline resampling and the global transition model has the promising discriminative capability in comparison with other ones.
引用
收藏
页码:7195 / 7220
页数:26
相关论文
共 16 条
[1]   A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking [J].
Arulampalam, MS ;
Maskell, S ;
Gordon, N ;
Clapp, T .
IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2002, 50 (02) :174-188
[2]   Kernel-based object tracking [J].
Comaniciu, D ;
Ramesh, V ;
Meer, P .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2003, 25 (05) :564-577
[3]   Comparison of resampling schemes for particle filtering [J].
Douc, R ;
Cappé, O ;
Moulines, E .
ISPA 2005: Proceedings of the 4th International Symposium on Image and Signal Processing and Analysis, 2005, :64-69
[4]   An Improvement on Resampling Algorithm of Particle Filters [J].
Fu, Xiaoyan ;
Jia, Yingmin .
IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2010, 58 (10) :5414-5420
[5]   Following a moving target - Monte Carlo inference for dynamic Bayesian models [J].
Gilks, WR ;
Berzuini, C .
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY, 2001, 63 :127-146
[6]   NOVEL-APPROACH TO NONLINEAR NON-GAUSSIAN BAYESIAN STATE ESTIMATION [J].
GORDON, NJ ;
SALMOND, DJ ;
SMITH, AFM .
IEE PROCEEDINGS-F RADAR AND SIGNAL PROCESSING, 1993, 140 (02) :107-113
[7]  
Hearn D., 1997, Computer graphics
[8]   CONDENSATION - Conditional density propagation for visual tracking [J].
Isard, M ;
Blake, A .
INTERNATIONAL JOURNAL OF COMPUTER VISION, 1998, 29 (01) :5-28
[9]   Gaussian particle filtering [J].
Kotecha, JH ;
Djuric, PA .
IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2003, 51 (10) :2592-2601
[10]   Improved particle filter for nonlinear system state [J].
Liang, J. ;
Peng, X. -Y. .
ELECTRONICS LETTERS, 2008, 44 (21) :1275-1276