Local smoothness for global optical flow

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

  • Lars Lau Rakêt
We consider the problem of estimating the “smoothness parameter” that controls the tradeoff between data fidelity and regularity in optical flow estimation. We start by reviewing the problem of global estimation using the Optimal Prediction Principle (OPP) by Zimmer et al. Inspired by this technique and work on local-global optical flow we propose a simple method for fusing optical flow estimates of different smoothness by evaluating interpolation quality locally by means of L1 block match on the corresponding set of gradient images. We illustrate the method in a setting where optical flows are estimated by a TV-L1 energy. On average this procedure reduces the average endpoint error by 15% over flows estimated using the OPP, and gives flow fields that are consistently better than the single best flows with a fixed smoothness parameter.
Original languageEnglish
Title of host publication2012 19th IEEE International Conference on Image Processing (ICIP)
Number of pages4
PublisherIEEE
Publication date2012
ISBN (Print)978-1-4673-2534-9
ISBN (Electronic)978-1-4673-2533-2
DOIs
Publication statusPublished - 2012
Event2012 19th IEEE International Conference on Image Processing - Orlando , United States
Duration: 30 Sep 20123 Oct 2012
Conference number: 19

Conference

Conference2012 19th IEEE International Conference on Image Processing
Nummer19
LandUnited States
ByOrlando
Periode30/09/201203/10/2012

ID: 38468477