Adaptive sparse vector tracking via online bayesian learning

Authors:
Yun Lei;Xiaoqing Ding;Shengjin Wang
Affiliations:
Electronic Engineering Department, Tsinghua University, Beijing, P.R. China;Electronic Engineering Department, Tsinghua University, Beijing, P.R. China;Electronic Engineering Department, Tsinghua University, Beijing, P.R. China
Venue:
IWICPAS'06 Proceedings of the 2006 Advances in Machine Vision, Image Processing, and Pattern Analysis international conference on Intelligent Computing in Pattern Analysis/Synthesis
Year:
2006

Citing 19
Cited 0

The nature of statistical learning theory

The nature of statistical learning theory
Mean Shift: A Robust Approach Toward Feature Space Analysis

IEEE Transactions on Pattern Analysis and Machine Intelligence
Quasi-Random Sampling for Condensation

ECCV '00 Proceedings of the 6th European Conference on Computer Vision-Part II
Kernel-Based Object Tracking

IEEE Transactions on Pattern Analysis and Machine Intelligence
Sparse bayesian learning and the relevance vector machine

The Journal of Machine Learning Research
On-Line Selection of Discriminative Tracking Features

ICCV '03 Proceedings of the Ninth IEEE International Conference on Computer Vision - Volume 2
Weighted and Robust Incremental Method for Subspace Learning

ICCV '03 Proceedings of the Ninth IEEE International Conference on Computer Vision - Volume 2
Robust Real-Time Face Detection

International Journal of Computer Vision
The Template Update Problem

IEEE Transactions on Pattern Analysis and Machine Intelligence
Distinctive Image Features from Scale-Invariant Keypoints

International Journal of Computer Vision
Ensemble Tracking

CVPR '05 Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Volume 2 - Volume 02
Online Selecting Discriminative Tracking Features Using Particle Filter

CVPR '05 Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Volume 2 - Volume 02
Sparse Bayesian Learning for Efficient Visual Tracking

IEEE Transactions on Pattern Analysis and Machine Intelligence
A Performance Evaluation of Local Descriptors

IEEE Transactions on Pattern Analysis and Machine Intelligence
Online Selection of Discriminative Tracking Features

IEEE Transactions on Pattern Analysis and Machine Intelligence
On-Line Density-Based Appearance Modeling for Object Tracking

ICCV '05 Proceedings of the Tenth IEEE International Conference on Computer Vision - Volume 2
A tutorial on particle filters for online nonlinear/non-GaussianBayesian tracking

IEEE Transactions on Signal Processing
Robust online appearance models for visual tracking

IEEE Transactions on Pattern Analysis and Machine Intelligence
Support vector tracking

IEEE Transactions on Pattern Analysis and Machine Intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

In order to construct a flexible representation for robust and efficient tracking, a novel real-time tracking method based on online learning is proposed in this paper. Under Bayesian framework, RVM is used to learn the log-likelihood ratio of the statistics of the interested object region to those of the nearby backgrounds. Then, the online selected sparse vectors by RVM are integrated to construct an adaptive representation of the tracked object. Meanwhile, the trained RVM classifier is embedded into particle filtering for tracking. To avoid distraction by the particles in background region, the extreme outlier model is incorporated to describe the posterior probability distribution of all particles. Subsequently, mean-shift clustering and EM algorithm are combined to estimate the posterior state of the tracked object. Experimental results over real-world sequences have shown that the proposed method can efficiently and effectively handle drastic illumination variation, partial occlusion and rapid changes in viewpoint, pose and scale.