%0 Conference Proceedings
%T A Confidence Measure for Assessing Optical Flow Accuracy in the Absence of Ground Truth
%A Patricia Marquez
%A Debora Gil
%A Aura Hernandez-Sabate
%B IEEE International Conference on Computer Vision – Workshops
%D 2011
%I IEEE
%C Barcelona (Spain)
%G English
%F Patricia Marquez2011
%O IAM; ADAS
%O exported from refbase (http://refbase.cvc.uab.es/show.php?record=1682), last updated on Wed, 16 Jan 2019 12:34:54 +0100
%X Optical flow is a valuable tool for motion analysis in autonomous navigation systems. A reliable application requires determining the accuracy of the computed optical flow. This is a main challenge given the absence of ground truth in real world sequences. This paper introduces a measure of optical flow accuracy for Lucas-Kanade based flows in terms of the numerical stability of the data-term. We call this measure optical flow condition number. A statistical analysis over ground-truth data show a good statistical correlation between the condition number and optical flow error. Experiments on driving sequences illustrate its potential for autonomous navigation systems.
%K IEEE International Conference on Computer Vision – Workshops
%U http://www.cvc.uab.es/adas/CVVT2011/
%U http://refbase.cvc.uab.es/files/MGH2011.pdf
%U http://dx.doi.org/10.1109/ICCVW.2011.6130499
%P 2042-2049