An algorithm to estimate the vertices of a tetrahedron from uniform random points inside (Q2413457)

Let \(T\) be the non-degenerate solid tetrahedron with (unknown) vertices \(A_1, A_2, A_3, A_4\) in \(\mathbb{R}^3\). Suppose it is possible to sample a sequence \(M_1, M_2, \ldots, M_n\) of independent uniformly distributed points on \(T\). This paper addresses the problem of estimation of \(A_1, A_2, A_3, A_4\) from \(M_1, \ldots, M_n\). The method goes by expressing the mixed moments \(\mathbb{E} [ X^i Y^j Z^k ]\), where \((X,Y,Z)\) is a uniform random vector on \(T\) expressed in Cartesian coordinates, in terms of the coordinates of \(A_1, A_2, A_3, A_4\), applying an inversion, and inserting in place of the mixed moments the empirical moments from the sample \(M_1, \ldots, M_n\). Results of Monte Carlo simulations are also presented.