Robust object tracking based on uncertainty factorization subspace constraints optical flow

Authors:
Yunshu Hou;Yanning Zhang;Rongchun Zhao
Affiliations:
College of Computer, Northwestern Polytechnical University, Xi’an, China;College of Computer, Northwestern Polytechnical University, Xi’an, China;College of Computer, Northwestern Polytechnical University, Xi’an, China
Venue:
CIS'05 Proceedings of the 2005 international conference on Computational Intelligence and Security - Volume Part II
Year:
2005

Citing 4
Cited 1

Efficient Region Tracking With Parametric Models of Geometry and Illumination

IEEE Transactions on Pattern Analysis and Machine Intelligence
A morphable model for the synthesis of 3D faces

Proceedings of the 26th annual conference on Computer graphics and interactive techniques
Active Appearance Models

IEEE Transactions on Pattern Analysis and Machine Intelligence
Factorization with Uncertainty

International Journal of Computer Vision

Robust shape-based head tracking

ACIVS'07 Proceedings of the 9th international conference on Advanced concepts for intelligent vision systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

The traditional methods of optical flow estimation have some problems, such as huge computation cost for the inverse of time-varying Hessian matrix, aperture phenomena for the points with 1D or little texture and drift phenomena with long sequences. A novel nonrigid object tracking algorithm based on inverse component uncertainty factorization subspace constraints optical flow is proposed in this paper, which resolves the above problems and achieves fast, robust and precise tracking. The idea of inverse Component is implemented in each recursive estimation procedure to make the algorithm fast. Uncertainty factorization is used to transform the optimization problem from a hyper-ellipse space to a hyper-sphere space. SVD is correspondingly performed to involve the subspace constraints. The proposed algorithm has been evaluated by both the standard test sequence and the consumer USB camera recorded sequence. The potential applications vary from articulated automation to structure from motion.