PT Unknown
AU Patricia Marquez
   Debora Gil
   Aura Hernandez-Sabate
TI A Confidence Measure for Assessing Optical Flow Accuracy in the Absence of Ground Truth
BT IEEE International Conference on Computer Vision – Workshops
PY 2011
BP 2042
EP 2049
DI 10.1109/ICCVW.2011.6130499
LA English
DE IEEE International Conference on Computer Vision – Workshops
AB 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.
PI Barcelona (Spain)
ER