This paper introduces an intermediate grouping phase whereby structures, meant to correspond to generic 3D junctions of planar and curved surfaces, are extracted from imperfect basic constant curvature contour primitives. This research work is part of a more generic project for detecting and describing generic high-level structures corresponding to 3D objects and/or parts of 3D objects in a single illuminance image of a cluttered scene. MAGNO (Multi-level Access to Generic Notable Objects), a computer vision system exploiting generic constraints at each of its hierarchical processing levels is at the heart of the project.
[1]
Jiri Matas,et al.
Junction detection using probabilistic relaxation
,
1993,
Image Vis. Comput..
[2]
Wolfgang Förstner,et al.
Polymorphic grouping for image segmentation
,
1995,
Proceedings of IEEE International Conference on Computer Vision.
[3]
David G. Lowe,et al.
Perceptual Organization and Visual Recognition
,
2012
.
[4]
Gérard G. Medioni,et al.
Perceptual grouping for generic recognition
,
2004,
International Journal of Computer Vision.
[5]
Jiri Matas,et al.
Low-level Grouping of Straight Line Segments
,
1991,
BMVC.