Title :
Extracting object silhouettes by perceptual edge grouping
Author_Institution :
Fac. of Comput. Sci., Dalhousie Univ., Halifax, NS, Canada
Abstract :
This paper presents a qualitative approach to extract object silhouettes based on principles of perceptual organization. A physical edge of object, i.e., silhouette, is modeled as a set of sequentially connected generic segments partitioned at perceptually significant points of the edge. Each segment consists of a series of proximate edge pixels which satisfy a descriptive property of the grouping. An object silhouette is reconstructed by combining the edge segments together
Keywords :
computational geometry; computer vision; edge detection; feature extraction; image reconstruction; optical tracking; computer vision; descriptive geometry; edge partition; feature extraction; generic segments; image reconstruction; object silhouettes; perceptual edge grouping; perceptual organization; tracking; Data mining; Humans; Image analysis; Image edge detection; Image segmentation; Joining processes; Polynomials; Psychology; Shape; Visual perception;
Conference_Titel :
Systems, Man, and Cybernetics, 1997. Computational Cybernetics and Simulation., 1997 IEEE International Conference on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-4053-1
DOI :
10.1109/ICSMC.1997.635296
