Window-Based Range Flow with an Isometry Constraint

This paper proposes a simple window-based range flow method which uses isometry of the observed surface as its primary matching constraint. The method uses feature points as anchoring references of the surface deformation. Given a set of matched features no other intensity information is used and hence the method can tolerate intensity changes over time. The range-flow equation is only required for a final verification step making the method robust to poor quality range images. This allows us to use the popular Point Grey Research Bumblebee 2 stereo-head to acquire our range data. The approach is shown to work well on two example scenes which capture non-rigid isometric and general deformations. The paper also presents experiments demonstrating the stability of the geodesic approximation employed in the isometry-based matching when the 3D point clouds are sparse.