The covariogram determines three-dimensional convex polytopes

TL;DR

This paper proves that the cross covariogram uniquely determines three-dimensional convex polytopes and certain convex polyhedral cones, settling a conjecture and advancing inverse geometric analysis.

Contribution

It establishes that the covariogram uniquely identifies 3D convex polytopes and specific cones, resolving a longstanding conjecture in convex geometry.

Findings

g_{K,K} determines 3D convex polytopes K

g_{K,L} determines K and L for certain convex cones

Addresses known counterexamples in higher dimensions

Abstract

The cross covariogram g_{K,L} of two convex sets K, L in R^n is the function which associates to each x in R^n the volume of the intersection of K with L+x. The problem of determining the sets from their covariogram is relevant in stochastic geometry, in probability and it is equivalent to a particular case of the phase retrieval problem in Fourier analysis. It is also relevant for the inverse problem of determining the atomic structure of a quasicrystal from its X-ray diffraction image. The two main results of this paper are that g_{K,K} determines three-dimensional convex polytopes K and that g_{K,L} determines both K and L when K and L are convex polyhedral cones satisfying certain assumptions. These results settle a conjecture of G. Matheron in the class of convex polytopes. Further results regard the known counterexamples in dimension n>=4. We also introduce and study the notion of synisothetic polytopes. This concept is related to the rearrangement of the faces of a convex polytope.