PCA based modification of SIFT-like methods for object class recognition

• Autorzy: Owczarek A.
• Języki publikacji: EN

Abstrakty

EN

This paper discusses a novel PCA based modification of standard SIFT and PCA-SIFT algorithms for the purpose of object class recognition. New descriptors intended to be simultaneously distinctive enough to describe the difference between features belonging to separate categories and general enough to capture the variations among features from the same class are proposed. The experimental results, gained for a test database, showing the reliability of introduced approach are presented.

Słowa kluczowe

EN

object recognition; principle component analysis; SIFT

• Wydawca: Foundation for Young Scientists
• Czasopismo: Challenges of Modern Technology
• Rocznik: 2011
• Tom: Vol. 2, no. 1
• Strony: 23--26
• Opis fizyczny: Bibliogr. 8 poz., tab., fot.

Twórcy

autor

Owczarek A.

• Institute of Electronics, Technical University of Lodz, 211/215 Wolczanska str., 90-924 Lodz

Bibliografia

• [1] Amit, Y., and D. Geman. ”A computational model for visual selection”. Neural Computation 11(7), 1999: 1691--1715.
• [2] Borenstein, E., and S. Ullman. “Class-specific, top--down segmentation”. Proceedings of ECCV 2002: 109--124.
• [3] Burl, M., M. Weber, and P. Perona. “A probabilistic approach to object recognition using local photometry and global geometry”. Proceedings of ECCV 1998: 628--641.
• [4] Schneiderman, H., and T. Kanade. “A statistical method for 3d object detection applied to faces and cars”. Proceedings of CVPR 2000.
• [5] Lowe, D.G. “Distinctive Image Features from Scale-Invariant Keypoints”. International Journal of Computer Vision 2004: 91–110.
• [6] Ke, Y., and R. Sukthankar. “PCA-SIFT: A More Distinctive Representation for Local Image Descriptors”. 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’04) 2 (2004): 506--513.
• [7] Joliff e, I.T. Principal Component Analysis. Springer-Verlag: 1986.
• [8] Turk, M., and A. Pentland. “Face recognition using eigenfaces”. Proceedings of Computer Vision and Pattern Recognition 1991: 586–591.