| Bayesian algorithms for simultaneous structure from motion estimation of multiple independently moving objects. | |
| | |
MedLine Citation:
|
PMID: 15646875 Owner: NLM Status: MEDLINE |
Abstract/OtherAbstract:
|
In this paper, the problem of simultaneous structure from motion estimation for multiple independently moving objects from a monocular image sequence is addressed. Two Bayesian algorithms are presented for solving this problem using the sequential importance sampling (SIS) technique. The empirical posterior distribution of object motion and feature separation parameters is approximated by weighted samples. The first algorithm addresses the problem when only two moving objects are present. A singular value decomposition (SVD)-based sample clustering algorithm is shown to be capable of separating samples related to different objects. A pair of SIS procedures is used to track the posterior distribution of the motion parameters. In the second algorithm, a balancing step is added into the SIS procedure to preserve samples of low weights so that all objects have enough samples to propagate empirical motion distributions. By using the proposed algorithms, the relative motions of all the moving objects with respect to the camera can be simultaneously estimated. Both algorithms have been tested on synthetic and real-image sequences. Improved results have been achieved. |
| | |
Authors:
|
Gang Qian; Rama Chellappa; Qinfen Zheng |
Related Documents
:
|
10673015 - Promoter sequences and algorithmical methods for identifying them. 17108385 - Correspondence-free determination of the affine fundamental matrix. 12850295 - Ordered subsets bayesian tomographic reconstruction using 2-d smoothing splines as priors. 12269575 - Deriving inherent optical properties from water color: a multiband quasi-analytical alg... 23292765 - Estimation of finite population duration distributions from longitudinal survey panels ... 10673015 - Promoter sequences and algorithmical methods for identifying them. |
Publication Detail:
|
Type: Comparative Study; Evaluation Studies; Journal Article; Research Support, U.S. Gov't, Non-P.H.S. |
Journal Detail:
|
Title: IEEE transactions on image processing : a publication of the IEEE Signal Processing Society Volume: 14 ISSN: 1057-7149 ISO Abbreviation: IEEE Trans Image Process Publication Date: 2005 Jan |
Date Detail:
|
Created Date: 2005-01-13 Completed Date: 2005-02-10 Revised Date: 2006-11-15 |
Medline Journal Info:
|
Nlm Unique ID: 9886191 Medline TA: IEEE Trans Image Process Country: United States |
Other Details:
|
Languages: eng Pagination: 94-109 Citation Subset: IM |
Affiliation:
|
Department of Electrical and Computer Engineering and the Institute for Advanced Computer Studies, University of Maryland, College Park, MD 20742, USA. gqian@cfar.umd.edu |
Export Citation:
|
APA/MLA Format Download EndNote Download BibTex |
| MeSH Terms | |
Descriptor/Qualifier:
|
Algorithms* Artificial Intelligence* Cluster Analysis Computer Graphics Computer Simulation Humans Image Enhancement / methods Image Interpretation, Computer-Assisted / methods* Imaging, Three-Dimensional / methods Information Storage and Retrieval / methods Models, Biological Models, Statistical Movement / physiology* Numerical Analysis, Computer-Assisted Pattern Recognition, Automated / methods* Reproducibility of Results Sensitivity and Specificity Signal Processing, Computer-Assisted Subtraction Technique* Video Recording / methods* Walking / physiology |
From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine
Previous Document: Statistical processing of large image sequences.
Next Document: Multidimensional, mapping-based complex wavelet transforms.