Symmetry is an effective geometric cue to facilitate conventional segmentation techniques on images of man-made environment. Based on three fundamental principles that summarize the relations between symmetry and perspective imaging, namely, structure from symmetry, symmetry hypothesis testing, and global symmetry testing, we develop a prototype system which is able to automatically segment symmetric objects in space from single 2D perspective images. The result of such a segmentation is a hierarchy of geometric primitives, called symmetry cells and complexes, whose 3D structure and pose are fully recovered. Such a geometrically meaningful segmentation may greatly facilitate applications such as feature matching and robot navigation.
Geometric segmentation of perspective images based on symmetry groups
Proceedings Ninth IEEE International Conference on Computer Vision ; 1251-1258 vol.2
2003-01-01
944846 byte
Conference paper
Electronic Resource
English
Geometric Segmentation of Perspective Images Based on Symmetry Groups
British Library Conference Proceedings | 2003
|Symmetry-based 3-D reconstruction from perspective images
British Library Online Contents | 2005
|Segmentation of Range Images as the Search for Geometric Parametric Models
British Library Online Contents | 1995
|Coupled Geometric and Texture Pde-Based Segmentation
British Library Conference Proceedings | 2005
|Coupled geometric and texture PDE-based segmentation
IEEE | 2005
|