Ordered over-relaxation based Langevin Monte Carlo sampling for visual tracking

Abstract Visual tracking is a fundamental research topic in computer vision community, which is of great importance in many application areas including augmented reality, traffic control, medical imaging and video editing. This paper presents an ordered over-relaxation Langevin Monte Carlo sampling (ORLMC) based tracking method within the Bayesian filtering framework, in which the traditional object state variable is augmented with an auxiliary momentum variable. At the proposal step, the proposal distribution is designed by simulation of the Hamiltonian dynamics. We first use the ordered over-relaxation method to draw the momentum variable which could suppress the random walk behavior in Gibbs sampling stage. Then, we leverage the gradient of the energy function of the posterior distribution to draw new samples with high acceptance ratio. The proposed tracking method could ensure that the tracker will not be trapped in local optimum of the state space. Experimental results show that the proposed tracking method successfully tracks the objects in different video sequences and outperforms several conventional methods.

[1]  Ling Shao,et al.  Recent advances and trends in visual tracking: A review , 2011, Neurocomputing.

[2]  Gareth O. Roberts,et al.  Examples of Adaptive MCMC , 2009 .

[3]  Jun Wang,et al.  Robust visual tracking by metric learning with weighted histogram representations , 2015, Neurocomputing.

[4]  Yee Whye Teh,et al.  Bayesian Learning via Stochastic Gradient Langevin Dynamics , 2011, ICML.

[5]  Shai Avidan,et al.  Locally Orderless Tracking , 2012, CVPR.

[6]  Frank Dellaert,et al.  MCMC-based particle filtering for tracking a variable number of interacting targets , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[7]  W. K. Hastings,et al.  Monte Carlo Sampling Methods Using Markov Chains and Their Applications , 1970 .

[8]  Yao Lu,et al.  Nearest neighbor field driven stochastic sampling for abrupt motion tracking , 2014, 2014 IEEE International Conference on Multimedia and Expo (ICME).

[9]  ShaoLing,et al.  Recent advances and trends in visual tracking , 2011 .

[10]  Jeffrey S. Rosenthal,et al.  Optimal Proposal Distributions and Adaptive MCMC , 2011 .

[11]  Michael Isard,et al.  CONDENSATION—Conditional Density Propagation for Visual Tracking , 1998, International Journal of Computer Vision.

[12]  Edilson de Aguiar,et al.  Visual tracking with VG-RAM Weightless Neural Networks , 2016, Neurocomputing.

[13]  Hans Driessen,et al.  Langevin Monte Carlo filtering for target tracking , 2015, 2015 18th International Conference on Information Fusion (Fusion).

[14]  Mingyu Lu,et al.  Robust particle tracker via Markov Chain Monte Carlo posterior sampling , 2013, Multimedia Tools and Applications.

[15]  Yuan Li,et al.  Tracking in Low Frame Rate Video: A Cascade Particle Filter with Discriminative Observers of Different Lifespans , 2007, CVPR.

[16]  Junseok Kwon,et al.  Visual tracking decomposition , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[17]  Yao Lu,et al.  Abrupt motion tracking via adaptive stochastic approximation Monte Carlo sampling , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[18]  Meng Wang,et al.  Robust visual tracking via multi-graph ranking , 2015, Neurocomputing.

[19]  Zhihan Lv,et al.  Multimodal Hand and Foot Gesture Interaction for Handheld Devices , 2014, TOMM.

[20]  Jian Zhang,et al.  Abrupt Motion Tracking via Nearest Neighbor Field Driven Stochastic Sampling , 2015, Neurocomputing.

[21]  Yuan Li,et al.  Tracking in Low Frame Rate Video: A Cascade Particle Filter with Discriminative Observers of Different Lifespans , 2007, 2007 IEEE Conference on Computer Vision and Pattern Recognition.

[22]  Xiaoqin Zhang,et al.  Contour tracking with abrupt motion , 2009, 2009 16th IEEE International Conference on Image Processing (ICIP).

[23]  Yuan Yan Tang,et al.  Multiview Hessian discriminative sparse coding for image annotation , 2013, Comput. Vis. Image Underst..

[24]  Weifeng Liu,et al.  Multiview Hessian Regularization for Image Annotation , 2013, IEEE Transactions on Image Processing.

[25]  Junseok Kwon,et al.  Wang-Landau Monte Carlo-Based Tracking Methods for Abrupt Motions , 2013, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[26]  Louahdi Khoudour,et al.  Robust visual tracking via MCMC-based particle filtering , 2012, 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[27]  Dongbing Gu,et al.  Abrupt motion tracking using a visual saliency embedded particle filter , 2014, Pattern Recognit..

[28]  Dan Schonfeld,et al.  Visual Tracking Using High-Order Particle Filtering , 2011, IEEE Signal Processing Letters.

[29]  Xuelong Li,et al.  Hessian Regularized Support Vector Machines for Mobile Image Annotation on the Cloud , 2013, IEEE Transactions on Multimedia.

[30]  Jiwen Lu,et al.  Abrupt Motion Tracking Via Intensively Adaptive Markov-Chain Monte Carlo Sampling , 2012, IEEE Transactions on Image Processing.

[31]  Alberto Del Bimbo,et al.  Particle filter-based visual tracking with a first order dynamic model and uncertainty adaptation , 2011, Comput. Vis. Image Underst..

[32]  Johan Dahlin,et al.  Particle metropolis hastings using Langevin dynamics , 2013, 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.

[33]  Junseok Kwon,et al.  Tracking of Abrupt Motion Using Wang-Landau Monte Carlo Estimation , 2008, ECCV.

[34]  Dapeng Tao,et al.  Online tracking based on efficient transductive learning with sample matching costs , 2016, Neurocomputing.

[35]  Patrick Pérez,et al.  Color-Based Probabilistic Tracking , 2002, ECCV.