With the advent of feature-based parametric models, many new design and analysis techniques have been developed to take advantage of the model’s features and design parameters. Examples of this are new grid generation techniques for Cartesian and overset methods. Unfortunately, many legacy geometries exist for which feature-based representations are not available. Proposed herein is a new technique for creating a parametric model from a legacy geometry that is defined in terms of a cloud of points that each lie on the aircraft’s surfaces. First, for the points associated with any component, a variant of the Levenberg-Marquardt gradient-based optimization method (LM) is used to find the set of model parameters that minimizes the least-square errors between the model and the points. The efficiency of the LM algorithm is greatly improved through the use of analytic geometric sensitivities and sparse matrix techniques. Second, for cases in which one does not know a priori the correspondences between points in the cloud and the aircraft’s components, an efficient initialization and classification algorithm is introduced. The technique is first explained with super-ellipse and isolated wing/fuselage configurations; then it is applied to aircraft configurations such as a glider. The accuracy and efficiency of these new techniques are demonstrated in both two and three dimensions, for configurations with both single and multiple components.