Smoothing of optical flow using robustified diffusion kernels

Date

2010

Authors

Doshi, A.
Bors, A.G.

Editors

Advisors

Journal Title

Journal ISSN

Volume Title

Type:

Journal article

Citation

Image and Vision Computing, 2010; 28(12):1575-1589

Statement of Responsibility

Conference Name

Abstract

This paper proposes a new optical flow smoothing methodology combining vector diffusion and robuststatistics. Vector smoothing using diffusion preserves moving object boundaries and the main motiondiscontinuities. According to a study provided in the paper, diffusion does not remove the outliers butspreads them out, introducing a bias in the neighbourhood. In this paper robust statistics operators such asthe median and alpha-trimmed mean are considered for robustifying the diffusion kernels. The robustdiffusion smoothing process is extended to 3-D lattices as well. The proposed algorithms are applied forsmoothing artificially generated vector fields as well as the optical flow estimated from image sequences.

School/Discipline

Dissertation Note

Provenance

Description

Data source: Figures & tables, https://doi.org/10.1016/j.imavis.2010.04.001

Access Status

Rights

Copyright 2010 Elsevier BV

License

Grant ID

Call number

Persistent link to this record