News for hackers. Stuff that matters.
Re:A new algorithm for finding a visual center of a polygon by Iv on Saturday May 20, 2017 @10:46PM (#14385449)
> While their approach is clever, it is not guaranteed to converge in a fixed number of iterations.

Are you sure about it? I struggle to find a case where the quadtree version of their algorithm would not converge.

Re:A new algorithm for finding a visual center of a polygon by opticalflow on Sunday May 21, 2017 @10:03AM (#14387345)
I didn't mean to imply that there were cases where it would Never converge, just that the number of iterations is not a-priori fixed or predictable.
Re:A new algorithm for finding a visual center of a polygon by Iv on Monday May 22, 2017 @01:08AM (#14391136)
My intuition is that the worst case scenario would be a fractal-like curve with multiple solutions (imagine a snake-like polygon following a Hilbert pattern). In that case the algorithm would need to explore all of the smallest squares, that gives us a higher bound on the number of iterations. It will be far less in most cases.

Would the delaunay-based algorithm be faster in the worst case?