On Symmetry and Multiple-View Geometry: Structure, Pose, and Calibration from a Single Image
International Journal of Computer Vision
| Hi-index | 0.00 |
Based on high-level geometric knowledge, especially symmetry,imposed upon objects in images, we demonstrate in this paper how toedit images in terms of correct 3-D shape and relationships of theobjects, without explicitly performing full 3-D reconstruction.Symmetry is proposed as the central notation that unifies bothconceptually and algorithmically all types geometric regularitiessuch as parallelism, orthoganality, and similarity. The methods areextremely simple, accurate, and easy to implement, and theydemonstrate the power of applying scene knowledge.