Segmentation and Recovery of SHGCs from a Real Intensity Image

We address the problem of scene segmentation and shape recovery from a single real intensity image. Solving this problem is central to obtaining 3-D scene descriptions in realistic applications where perfect data cannot be obtained and only one image is available. The method we propose addresses a large class of generic shapes, namely straight homogeneous generalized cylinders (SHGCs). It consists of the derivation and use of their geometric projective properties in a multi-level grouping approach. We describe an implemented and working system that detects and recovers full SHGC descriptions in the presence of image imperfections such as broken contours, surface markings, shadows and occlusion. We demonstrate our method on complex real images.

[1]  Vishvjit S. Nalwa.  Line-Drawing Interpretation: Bilateral Symmetry , 1989, IEEE Trans. Pattern Anal. Mach. Intell..

[2]  David A. Forsyth,et al.  Extracting projective structure from single perspective views of 3D point sets , 1993, 1993 (4th) International Conference on Computer Vision.

[3]  Philippe Saint-Marc,et al.  B-Spline Contour Representation and Symmetry Detection , 1990, ECCV.

[4]  R. Nevatia,et al.  Quasi-invariant properties and 3-D shape recovery of non-straight, non-constant generalized cylinders , 1993, Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[5]  T. Kanade,et al.  The Theory of Straight Homogeneous Generalized Cylinders , 1983 .

[6]  Ramakant Nevatia,et al.  Description and Recognition of Curved Objects , 1977, Artif. Intell..

[7]  David A. Forsyth,et al.  Efficient recognition of rotationally symmetric surfaces and straight homogeneous generalized cylinders , 1993, Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[8]  Jean Ponce,et al.  Invariant Properties of Straight Homogeneous Generalized Cylinders and Their Contours , 1989, IEEE Trans. Pattern Anal. Mach. Intell..

[9]  Ramakant Nevatia,et al.  Descriptions of complex objects from incomplete and imperfect data , 1989 .

[10]  Minoru Asada,et al.  A qualitative approach to quantitative recovery of SHGCs shape and pose from shading and contour , 1993, Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[11]  Michel Dhome,et al.  Inverse Perspective Transform Using Zero-Curvature Contour Points: Application to the Localization of Some Generalized Cylinders from a Single View , 1991, IEEE Trans. Pattern Anal. Mach. Intell..

[12]  Ramakant Nevatia,et al.  Perception of 3-D Surfaces from 2-D Contours , 1993, IEEE Trans. Pattern Anal. Mach. Intell..

[13]  R. Nevatia,et al.  Perceptual Organization for Scene Segmentation and Description , 1992, IEEE Trans. Pattern Anal. Mach. Intell..

[14]  I. Biederman.  Recognition-by-components: a theory of human image understanding. , 1987, Psychological review.

[15]  Ramakant Nevatia,et al.  Using Invariance and Quasi-Invariance for the Segmentation and Recovery of Curved Objects , 1993, Applications of Invariance in Computer Vision.

[16]  Michel Dhome,et al.  Recovering the scaling function of a SHGC from a single perspective view , 1992, Proceedings 1992 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[17]  Robert Bergevin,et al.  Generic Object Recognition: Building and Matching Coarse Descriptions from Line Drawings , 1993, IEEE Trans. Pattern Anal. Mach. Intell..

[18]  Thomas O. Binford,et al.  Inferring Surfaces from Images , 1981, Artif. Intell..

[19]  Rodney A. Brooks,et al.  Model-Based Three-Dimensional Interpretations of Two-Dimensional Images , 1981, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[20]  T. Binford,et al.  Finding and recovering SHGC objects in an edge image , 1993 .