Name
Abstract | ||
|---|---|---|
Existing approaches to template-based visual tracking, in which the objective is to continuously estimate the spatial transformation parameters of an object template over video frames, have primarily been based on deterministic optimization, which as is well-known can result in convergence to local optima. To overcome this limitation of the deterministic optimization approach, in this paper we present a novel particle filtering approach to template-based visual tracking. We formulate the problem as a particle filtering problem on matrix Lie groups, specifically the Special Linear group SL(3) and the two-dimensional affine group Aff (2). Computational performance and robustness are enhanced through a number of features: (i) Gaussian importance functions on the groups are iteratively constructed via local linearization; (ii) the inverse formulation of the Jacobian calculation is used; (iii) template resizing is performed; and (iv) parent-child particles are developed and used. Extensive experimental results using challenging video sequences demonstrate the enhanced performance and robustness of our particle filtering-based approach to template-based visual tracking. We also show that our approach outperforms several state-of-the-art template-based visual tracking methods via experiments using the publicly available benchmark dataset. |
Year | DOI | Venue |
|---|---|---|
2014 | 313E1DE7-71E5-49CD-B729-9E208589DB7B | IEEE Trans. Pattern Anal. Mach. Intell. |
Keywords | Field | DocType |
3d special linear group,lie groups,video signal processing,particle filtering (numerical methods),visual tracking,video frames,gaussian importance functions,video sequences,deterministic optimization approach,particle filtering,parameter estimation,spatial transformation parameter estimation,sl(3),computational geometry,object template,affine group,gaussian importance function,geometric particle filter,computational performance enhancement,matrix inversion,special linear group,jacobian matrices,object tracking,image sequences,local linearization,particle filtering approach,parent-child particles,jacobian calculation inverse formulation,2d affine group,matrix lie groups,lie group,computational robustness enhancement,template-based visual tracking,iterative methods,template resizing,approximation algorithms,visualization,mathematical model,tracking,algebra | Computer vision,Pattern recognition,Jacobian matrix and determinant,Computer science,Local optimum,Computational geometry,Particle filter,Robustness (computer science),Gaussian,Video tracking,Artificial intelligence,Linearization | Journal |
Volume | Issue | ISSN |
36 | 4 | 1939-3539 |
Citations | PageRank | References |
10 | 0.47 | 44 |
Authors | ||
4 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Junghyun Kwon | 1 | 128 | 8.28 |
| Hee Seok Lee | 2 | 65 | 3.79 |
| Frank Chongwoo Park | 3 | 377 | 61.92 |
| Kyoung Mu Lee | 4 | 3228 | 153.84 |