Smooth determinantal varieties and critical loci in multiview geometry (Q2164950)

Photos of static three-dimensional scenes taken from pinhole cameras are usually modelled by linear projections from \(\mathbb P^3\) to \(\mathbb P^2\). More generally, it is often the case that one studies linear projections from \(\mathbb P^k\) to \(\mathbb P^h\). The reconstruction problem is the following: given a set of points in \(\mathbb P^k\) with unknown coordinates, and \(n\) images of this set in \(n\) target spaces \(\mathbb P^{hi}\) taken from unknown ``cameras,'' can we recover the positions of the cameras and the original set of points? Given \(n\) sufficiently large one would expect to be able to do this. Nevertheless, there exist sets of points for which the projective reconstruction fails. These configurations of points are called critical, meaning that there exist other non projectively equivalent sets of points and cameras that give the same images. Critical loci turn out to be determinantal algebraic varieties. In this paper the authors determine and classify all the smooth critical loci and show that they are classical projective varieties. That is, they determine under what assumptions the critical locus is a smooth variety, and they provide a complete and effective classification of smooth varieties that can be critical.