THIS PAGE INCORPORATES CASCADING STYLE SHEET! TO VIEW PAGE CORRECTLY PLEASE UPDATE TO A CSS COMPLIANT BROWSER
 |
Flexible Flow for 3D Non-Rigid Tracking and Shape Recovery |
pdf 951 KB
ps.gz 1.26 MB
ps 3.52 MB
@InProceedings{sanders:OD-Theory:AAAI02,
author = {Mathew Brand and Rahul Bhotika},
title = {Flexible Flow for 3D Non-Rigid Tracking and Shape Recovery},
booktitle = {Proc. {AAAI02} Conf. on Computer Vision and Pattern Recognition {CVPR2001"},},
year = {2001},
address = {Kauai, Hawaii},
month = {Dec},
publisher = {{IEEE} Computer Society Press},
abstract = {We introduce linear methods for model-based tracking of nonrigid 3D objects and for acquiring such models from video. 3D motions and flexions are calculated directly from image intensities without information-lossy intermediate results. Measurement uncertainty is quantified and fully propagated through the inverse model to yeild posterior mean(PM) and/or mode (MAP) pose estimates. A Bayesian framework manages uncertainty, accommodates priors, and gives confidence meansures. We obtain highly accurate and robust closed-form estimators by minimizing information loss from non-reversible (inner-product and least-squares) operations, and, when unavoidable, performing such operations with the appropriate error norm. For model acquisition, we show how to refine a crude or gneric model to fit the video subjuct. We demonstrate with tracking, model refinement, and super-resolution texture lifting from low-quality low-resolution video. },
}