Skip to main content

Timeline for Closest point on Bézier spline

Current License: CC BY-SA 4.0

14 events
when toggle format what by license comment
Jan 15 at 1:07 comment added SteveWithamDuplicate Oops, crossing sticks can get false positives but not false negatives. To make sure a positive is true, enclose each curve segment in a triangle with one side being the "stick" between the curve segment's end points. If the sticks pierce all the way thru each other's triangle, there's an intersection. If the sticks cross and stick A's end is inside triangle B, check whether stick a crosses curve B. Erm, easy peasy.
Jan 14 at 23:58 comment added SteveWithamDuplicate I don't see a mention here of the fact that cubic Bezier splines can intersect themselves (at most once I think). Looks like if one marks values of t where either x(t) or y(t) hits a local minimum or maximum, then draws a stick figure of the curve, one can find whether and which two monotonic sections of curve intersect.
Nov 27, 2022 at 15:01 history edited Martin Sleziak CC BY-SA 4.0
http -> https (the question was bumped anyway)
Nov 24, 2022 at 0:20 history edited LSpice CC BY-SA 4.0
Bezier -> Bézier, while this is on the front page
Nov 24, 2022 at 0:02 answer added Ken Seehart timeline score: 0
Mar 3, 2021 at 7:37 history edited gmvh CC BY-SA 4.0
Improved formatting, replaced inappropriate tags with appropriate ones
Feb 28, 2015 at 16:50 comment added Federico Poloni Which algorithm does it use, for those of us who do not have 45$ to spare?
Feb 28, 2015 at 16:06 comment added Varaughe If you want to see an implementation of calculating the distance from one point to a Bezier Curve(the closest point) , you can check out the "Runtime Curve Editor" assetstore.unity3d.com/en/#!/content/11835 , that's an Unity package(you perhaps need to install Unity) ,all the code is available,is C#,but math is the same, the package is doing much more than just calculating that distance(projection) , the price of package is 45$.
Dec 15, 2009 at 21:04 vote accept David Rutten
Dec 15, 2009 at 15:44 answer added Greg Kuperberg timeline score: 14
Dec 15, 2009 at 15:32 comment added David Rutten Thanks lhf. Suppose I'll stick with my current solution then.
Dec 15, 2009 at 15:23 comment added lhf A reliable and fast numerical algorithm is your best bet. Analytic solutions, even if they exist (which I doubt in this case), are not always the best choice. Even finding the roots of a cubic polynomial with Cardano's formula is messy and numerically unstable.
Dec 15, 2009 at 13:03 comment added Andrew Stacey You may have more luck with this kind of question in the places mentioned in the FAQ: mathoverflow.net/faq#whatnot
Dec 15, 2009 at 13:00 history asked David Rutten CC BY-SA 2.5