Projective and illumination invariant representation of disjoint shapes

We describe a projectively invariant representation of disjoint contour groups, which is suitable for shape-based retrieval from an image database. It consists of simultaneous polar reparametrization of multiple curves where an invariant point is used as the origin. For each ray orientation, a cross-ratio of its intersections with other curves is taken as a value associated to the radius. With respect to other methods, this representation is less reliant on single curve properties, both for the construction of the projective basis and for calculating the signature. It is, therefore, more robust to contour gaps and image noise and is better suited to describing complex planar shapes defined by multiple disjoint curves. Moreover, we show that illumination invariance fits well within the proposed framework and can easily be introduced in the representation in order to make it more appropriate for shape-based retrieval. Experiments are reported on a database of real trademarks.

[1]  Kenichi Kanatani Computational cross ratio for computer vision , 1994 .

[2]  David G. Lowe,et al.  Perceptual Organization and Visual Recognition , 2012 .

[3]  Thierry Pun,et al.  Statistical structuring of pictorial databases for content-based image retrieval systems , 1996, Pattern Recognit. Lett..

[4]  Arnold W. M. Smeulders,et al.  Image Databases and Multi-Media Search , 1998, Image Databases and Multi-Media Search.

[5]  Sunil Arya,et al.  An optimal algorithm for approximate nearest neighbor searching fixed dimensions , 1998, JACM.

[6]  Alex Pentland,et al.  Photobook: tools for content-based manipulation of image databases , 1994, Electronic Imaging.

[7]  Thierry Pun,et al.  Planar Shape Databases with Affine Invariant Search , 1998, Image Databases and Multi-Media Search.

[8]  Fernand S. Cohen,et al.  Part I: Modeling Image Curves Using Invariant 3-D Object Curve Models-A Path to 3-D Recognition and Shape Estimation from Image Contours , 1994, IEEE Trans. Pattern Anal. Mach. Intell..

[9]  Thierry Pun,et al.  Correspondence analysis and hierarchical indexing for content-based image retrieval , 1996, Proceedings of 3rd IEEE International Conference on Image Processing.

[10]  Patrick Gros,et al.  Geometric interpretation of joint conic invariants , 1992 .

[11]  Ehud Rivlin,et al.  Local Invariants For Recognition , 1995, IEEE Trans. Pattern Anal. Mach. Intell..

[12]  Aaron Heller,et al.  An experimental evaluation of projective invariants , 1992 .

[13]  Arnold W. M. Smeulders,et al.  A comparative study of several color models for color image invariant retrieval , 1996 .

[14]  Richard I. Hartley,et al.  Projective Reconstruction and Invariants from Multiple Images , 1994, IEEE Trans. Pattern Anal. Mach. Intell..

[15]  Andrew Zisserman,et al.  Geometric invariance in computer vision , 1992 .

[16]  Andrew Zisserman,et al.  Recognizing general curved objects efficiently , 1992 .

[17]  David F. Rogers,et al.  Mathematical elements for computer graphics , 1976 .

[18]  T. Pun,et al.  Recognition of planar objects over complex backgrounds using line invariants and relevance measures , 1995 .

[19]  Dragutin Petkovic,et al.  Query by Image and Video Content: The QBIC System , 1995, Computer.

[20]  Toshikazu Kato,et al.  Query by Visual Example - Content based Image Retrieval , 1992, EDBT.

[21]  Graham D. Finlayson,et al.  Color constancy in diagonal chromaticity space , 1995, Proceedings of IEEE International Conference on Computer Vision.

[22]  Thomas H. Reiss,et al.  Recognizing Planar Objects Using Invariant Image Features , 1993, Lecture Notes in Computer Science.