Visual reconstruction
3-D multiview object representation for model-based object recognition
Pattern Recognition
Parallel and Deterministic Algorithms from MRFs: Surface Reconstruction
IEEE Transactions on Pattern Analysis and Machine Intelligence
Object Identification and Pose Estimation for Automatic Manipulation
RobVis '01 Proceedings of the International Workshop on Robot Vision
Hi-index | 0.00 |
Several methods have been proposed for building 3-D object models from multiple range images mainly on the basis of the deformable models and pseudo-dynamics. However, the patches are usually placed crossing over the edge regions, and the surface orientation discontinuities cannot be properly preserved as a result. This paper proposes a new 3-D object modeling method that generates refined triangular meshes while preserving the surface discontinuities. We formulate the process into a problem of minimizing an energy functional by using the idea of line processes and MRFs. Moreover, we show that next-to-optimal solutions can be obtained by utilizing a continuation method.