Drift robust non-rigid optical flow enhancement for long sequences
File(s)
Accepted version
Author(s)
Li, W
Cosker, D
Brown, M
Type
Journal Article
Abstract
It is hard to densely track a nonrigid object in long term, which is a fundamental research issue in the computer vision community. This task often relies on estimating pairwise correspondences between images over time where the error is accumulated and leads to a drift. In this paper, we introduce a novel optimisation framework with an Anchor Patch constraint. It is supposed to significantly reduce overall errors given long sequences containing nonrigidly deformable objects. Our framework can be applied to any dense tracking algorithm, e.g. optical flow. We demonstrate the success of our approach by showing significant error reduction on 6 popular optical flow algorithms applied to a range of realworld nonrigid benchmarks. We also provide quantitative analysis of our approach given synthetic occlusions and image noise.
Editor(s)
Lv, Z
Date Issued
2016-10-13
Date Acceptance
2016-10-01
Citation
Journal of Intelligent and Fuzzy Systems, 2016, 31 (5), pp.2583-2595
ISSN
1064-1246
Publisher
IOS Press
Start Page
2583
End Page
2595
Journal / Book Title
Journal of Intelligent and Fuzzy Systems
Volume
31
Issue
5
Copyright Statement
© 2016 – IOS Press and the authors. All rights reserved
Subjects
cs.CV
0801 Artificial Intelligence And Image Processing
1702 Cognitive Science
Artificial Intelligence & Image Processing
Publication Status
Published