Semi-local variational optical flow estimation

Abstract : Global variational methods for optical flow estimation usually suffer from an over-smoothing effect. We propose a semilocal estimation framework designed to integrate and improve any variational method. The idea is to implicitly segment the minimization domain into coherently moving windows. In a first time, local variational estimations are performed in overlapping candidate square regions. Then, a global discrete optimization, non subject to the over-smoothing introduced by variational approaches, selects the optimal window for each pixel. Experimental results show an increasing of the sharpness of discontinuities and a significant improvement of global registration errors compared to the results of the baseline global variational method.
Document type :
Conference papers
Complete list of metadatas

Cited literature [9 references]  Display  Hide  Download

https://hal.inria.fr/hal-00763806
Contributor : Charles Kervrann <>
Submitted on : Tuesday, December 11, 2012 - 3:11:33 PM
Last modification on : Wednesday, April 11, 2018 - 1:54:45 AM
Long-term archiving on : Tuesday, March 12, 2013 - 10:25:29 AM

File

DFortun_ICIP12.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00763806, version 1

Collections

Citation

Denis Fortun, Charles Kervrann. Semi-local variational optical flow estimation. IEEE International Conference on Image Processing, ICIP 2012, Sep 2012, Orlando, FL, United States. ⟨hal-00763806⟩

Share

Metrics

Record views

255

Files downloads

218