Abstract | ||
---|---|---|
We propose a principled framework to model persistent motion in dynamic scenes. In contrast to previous efforts on object tracking and optical flow estimation that focus on local motion, we primarily aim at inferring a global model of persistent and collective dynamics. With this in mind, we first introduce the concept of geometric flow that describes motion simultaneously over space and time, and derive a vector space representation based on Lie algebra. We then extend it to model complex motion by combining multiple flows in a geometrically consistent manner. Taking advantage of the linear nature of this representation, we formulate a stochastic flow model, and incorporate a Gaussian process to capture the spatial coherence more effectively. This model leads to an efficient and robust algorithm that can integrate both point pairs and frame differences in motion estimation. We conducted experiments on different types of videos. The results clearly demonstrate that the proposed approach is effective in modeling persistent motion. |
Year | DOI | Venue |
---|---|---|
2010 | 10.1109/CVPR.2010.5539848 | computer vision and pattern recognition |
Keywords | Field | DocType |
persistent motion modeling,lie algebra,motion estimation,robust algorithm,gaussian process,lie algebras,object tracking,persistent motion estimation,tracking,gaussian processes,vector space representation,geometric flow,optical flow estimation | Multidisciplinary approach,Operations research,Wireless sensor network,Geography | Conference |
Volume | Issue | ISSN |
2010 | 1 | 1063-6919 |
ISBN | Citations | PageRank |
978-1-4244-6984-0 | 10 | 0.52 |
References | Authors | |
6 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Dahua Lin | 1 | 1117 | 72.62 |
W. E. L. Grimson | 2 | 11451 | 2002.95 |
John W. Fisher III | 3 | 878 | 74.44 |