You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
According to "Adaptive Inliers Criterion" section of your paper, support inliers distance to initial ellipse candidate should be less than some threshold. However, function calcEllipseParametersAndValidate does not perform that check and function d_rosin is defined but not used. Therefore implementation allows for support inliers which are far from ellipse and still satsify regional limitation like some arc inside larger ellipse.
Am I missing something or this is on your roadmap?
The text was updated successfully, but these errors were encountered:
Hi, I also found this issue. But, I found that, in the function calcEllipseParametersAndValidate, the points used for "Adaptive Inliers Criterion" are the inliers of each LS rectangle in two LSs group. Maybe the author thought that the distance from these points to the current ellipse must meet the d_rosin-conition?
According to "Adaptive Inliers Criterion" section of your paper, support inliers distance to initial ellipse candidate should be less than some threshold. However, function calcEllipseParametersAndValidate does not perform that check and function d_rosin is defined but not used. Therefore implementation allows for support inliers which are far from ellipse and still satsify regional limitation like some arc inside larger ellipse.
Am I missing something or this is on your roadmap?
The text was updated successfully, but these errors were encountered: