As collaborative trackers based on traditional fusion strategy has poor robustness in complex environments
a novel adaptive interactive fusion tracking strategy based on the online updated transition probability matrix in a multiple model particle filter framework was proposed. Firstly
an iterative updating equation was obtained based on minimum mean square error estimation method based on the Bayes theory. Then
the numerical solution of the iterative equation was obtained by numerical integration algorithm. Finally
with the updated TPM and re-sampling technology
the adaptive interaction of prior state distributions for different trackers was achieved to guarantee the target state of transmitted particles with larger weights. Tracking experiments were performed in complex environments. The results demonstrate that the proposed adaptive interactive fusion strategy improves the correction function for Particle prior state and effectively avoids the 'tracking drifting' problem from error accumulation. So
the robustness of proposed collaborative tracker is more better than those single trackers or collaborative trackers based other fusion strategy.
关键词
Keywords
references
NUMMIARO K, KOLLER-MEIER E, VAN GOOL L. An adaptive color-based particle filter [J]. Image and Vision Computing, 2003, 21(1):99-110.
ZHUANG B H, LU H C, XIAO Z Y, et al.. Visual tracking via discriminative sparse similarity map [J]. IEEE Transactions on Image Processing, 2014, 23(4):1872-1881.
WANG Q, JIN X F.Vehicle detecting and tracking method in complex environments [J]. Chinese Journal of Liquid Crystals and Displays, 2016, 31(5):511-517. (in Chinese)
CUI X W, WU Q ZH, JIANG P, et al.. Affine-invariant target tracking based on subspace representation [J]. Infrared and Laser Engineering, 2015, 44(2):769-774. (in Chinese)
XING Y L, LI A H, CUI ZH G, et al.. Moving target tracking algorithm based on improved Kernelized correlation filter [J]. Infrared and Laser Engineering, 2016, 45(S1):214-221. (in Chinese)
ZHAO Y F.Improved mean-shift algorithm combined with adaptive kernel function [J]. Chinese Journal of Liquid Crystals and Displays, 2016, 31(12):1143-1148. (in Chinese)
CEHOVIN L, KRISTAN M, LEONARDIS A. Robust visual tracking using an adaptive coupled-layer visual model [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(4):941-953.
ZHONG W, LU H C, YANG M H. Robust object tracking via sparsity-based collaborative model [C]. 2012 IEEE Conference on Computer Vision and Pattern Recognition, IEEE, 2012:1838-1845.
GAO Y, JI R R, ZHANG L F, et al.. Symbiotic tracker ensemble toward a unified tracking framework [J]. IEEE Transactions on Circuits and Systems for Video Technology, 2014, 24(7):1122-1131.
BIRESAW T A, CAVALLARO A, REGAZZONI C S. Tracker-level fusion for robust Bayesian visual tracking [J]. IEEE Transactions on Circuits and Systems for Video Technology, 2015, 25(5):776-789.
JILKOV V P, LI X R. Online Bayesian estimation of transition probabilities for Markovian jump systems [J]. IEEE Transactions on Signal Processing, 2004, 52(6):1620-1630.
SANMIGUEL J C, CAVALLARO A, MARTINEZ J M. Adaptive online performance evaluation of video trackers [J]. IEEE Transactions on Image Processing, 2012, 21(5):2812-2823.
WU Y, LIM J, YANG M H. Object tracking benchmark [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37(9):1834-1848.
ROSS D A, LIM J, LIN R S, et al.. Incremental learning for robust visual tracking [J]. International Journal of Computer Vision, 2008, 77(1-3):125-141.
BABENKO B, YANG M H, BELONGIE S. Visual tracking with online multiple instance learning [C]. IEEE Conference on Computer Vision and Pattern Recognition, IEEE, 2009:983-990.