Matching and Pose Estimation of Noisy, Partial and Planar B-Rep Models

Three dimensional models are represented in various ways. One possibility are boundary representation (B-Rep) models, which contain geometric and topological information. This makes B-Reps suitable for tasks that need an explicit algebraic representation of the surface, e.g. numerical optimization or simulation. Reconstructing a B-Rep model of an object or the environment often is a tedious task needing much manual intervention. An intuitive way is to use a hand-held depth camera and perform a real-time reconstruction. In the domain of robotics, we previously presented a system [18] that is able to incrementally reconstruct a planar B-Rep model from a stream of organized point clouds acquired from a robot mounted camera. Since the acquisition poses are known, this approach can not directly be used in a setup with a hand-held camera. The contribution of this work is a new method for matching planar B-Rep models and estimating their relative pose. In particular, the input models can be noisy and incomplete containing very few geometric features like corners, which is likely to occur when a model from only one viewpoint is processed. In combination with our previous work, we show that our approach can be used to build a simultaneous location and mapping (SLAM) system for an easy reconstruction of planar B-Rep models.