Pattern Recognition

In this paper we present SIMBA, a content based image retrieval system performing queries based on image appearance. We consider absolute object positions irrelevant for image similarity here and therefore propose to use invariant features. Based on a general construction method (integration over the transformation group), we derive invariant feature histograms that catch different cues of image content: features that are strongly influenced by color and textural features that are robust to illumination changes. By a weighted combination of these features the user can adapt the similarity measure according to his needs, thus improving the retrieval results considerably. The feature extraction does not require any manual interaction, so that it might be used for fully automatic annotation in heavily fluctuating image databases.

[1]  Gregory D. Hager,et al.  Efficient Region Tracking With Parametric Models of Geometry and Illumination , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[2]  Nassir Navab,et al.  E-commerce direct marketing using augmented reality , 2000, 2000 IEEE International Conference on Multimedia and Expo. ICME2000. Proceedings. Latest Advances in the Fast Changing World of Multimedia (Cat. No.00TH8532).

[3]  Hans Burkhardt,et al.  Invariant Feature Histograms for Texture Classification , 1998 .

[4]  Hans Burkhardt,et al.  AUTOMATIC DETECTION OF ERRORS ON TEXTURES USING INVARIANT GREY SCALE FEATURES AND POLYNOMIAL CLASSIFIERS , 2000 .

[5]  Rama Chellappa,et al.  Simultaneous tracking and verification via sequential posterior estimation , 2000, Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662).

[6]  R. Vanderbei.  LOQO:an interior point code for quadratic programming , 1999 .

[7]  Gerald Sommer,et al.  Affine real-time face tracking using Gabor wavelet networks , 2000, Proceedings 15th International Conference on Pattern Recognition. ICPR-2000.

[8]  David Haussler,et al.  Probabilistic kernel regression models , 1999, AISTATS.

[9]  Nassir Navab,et al.  Single point active alignment method (SPAAM) for optical see-through HMD calibration for AR , 2000, Proceedings IEEE and ACM International Symposium on Augmented Reality (ISAR 2000).

[10]  Hanns Schulz-Mirbach,et al.  Invariant Features for Gray Scale Images , 1995, DAGM-Symposium.

[11]  Joachim M. Buhmann,et al.  Empirical evaluation of dissimilarity measures for color and texture , 1999, Proceedings of the Seventh IEEE International Conference on Computer Vision.

[12]  Dorin Comaniciu,et al.  Real-time tracking of non-rigid objects using mean shift , 2000, Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662).

[13]  Nassir Navab,et al.  Taking AR into large scale industrial environments: navigation and information access with mobile computers , 2001, Proceedings IEEE and ACM International Symposium on Augmented Reality.

[14]  Matti Pietikäinen,et al.  A comparative study of texture measures with classification based on featured distributions , 1996, Pattern Recognit..

[15]  William H. Press,et al.  Numerical recipes in C , 2002 .

[16]  N. Navab,et al.  Tracking and pose estimation for computer assisted localization in industrial environments , 2000, Proceedings Fifth IEEE Workshop on Applications of Computer Vision.

[17]  Nassir Navab,et al.  Fusion of color, shading and boundary information for factory pipe segmentation , 2000, Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662).

[18]  Gerald Sommer,et al.  Gabor Wavelet Networks for Object Representation , 2000, Theoretical Foundations of Computer Vision.

[19]  Yakup Genc,et al.  Augmented workspace: designing an AR testbed , 2000, Proceedings IEEE and ACM International Symposium on Augmented Reality (ISAR 2000).

[20]  Gerald Sommer,et al.  Efficient Head Pose Estimation with Gabor Wavelet Networks , 2000, BMVC.