Many visualization tasks face the problem of rendering images of data sets which are represented by a (usually huge) set of points in space, without any known connectivity between the points.

We present a method to approximate these data sets at multiple levels of resolutions by using hierarchies of triangle meshes as simple surface representations.

Applications for this algorithm are surface reconstruction from unorganized points, and, surprisingly, image compression.

Snacks will be provided.

See Conundrum Talks for more information about this series.