Object Recognition using Local Affine Frames on Distinguished Regions

A novel approach to appearance based object recognition is introduced. The proposed method, based on matching of local image features, reliably recognises objects under very different viewing conditions. First, distinguished regions of data-dependent shape are robustly detected. On these regions, local affine frames are established using several affine invariant constructions. Direct comparison of photometrically normalised colour intensities in local, geometrically aligned frames results in a matching scheme that is invariant to piecewise-affine image deformations, but still remains very discriminative. The potential of the approach is experimentally verified on COIL-100 and SOIL-47 ‐ publicly available image databases. On SOIL-47, 100% recognition rate is achieved for single training view per object. On COIL-100, 99.9% recognition rate is obtained for 18 training views per object. Robustness to severe occlusions is demonstrated by only a moderate decrease of recognition performance in an experiment where half of each test image is erased.

[1]  Yehezkel Lamdan,et al.  Geometric Hashing: A General And Efficient Model-based Recognition Scheme , 1988, [1988 Proceedings] Second International Conference on Computer Vision.

[2]  Gérard G. Medioni,et al.  Finding Waldo, or focus of attention using local color information , 1993, Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[3]  Emanuele Trucco,et al.  Geometric Invariance in Computer Vision , 1995 .

[4]  Katsushi Ikeuchi,et al.  Detectability, Uniqueness, and Reliability of Eigen Windows for Stable Verification of Partially Occluded Objects , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

[5]  Cordelia Schmid,et al.  Local Grayvalue Invariants for Image Retrieval , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

[6]  Jan Flusser,et al.  Convex Layers: A New Tool for Recognition of Projectively Deformed Point Sets , 1999, CAIP.

[7]  Narendra Ahuja,et al.  Learning to Recognize 3D Objects with SNoW , 2000, ECCV.

[8]  Luc Van Gool,et al.  Wide Baseline Stereo Matching based on Local, Affinely Invariant Regions , 2000, BMVC.

[9]  Cordelia Schmid,et al.  Indexing Based on Scale Invariant Interest Points , 2001, ICCV.

[10]  Josef Kittler,et al.  On Matching Algorithms for the Recognition of Objects in Cluttered Background , 2001, IWVF.

[11]  Stepán Obdrzálek,et al.  Local Affine Frames for Image Retrieval , 2002, CIVR.

[12]  Heinrich Niemann,et al.  A Spin-Glass Markov Random Field for 3-D Object Recognition , 2002 .

[13]  Michael J. Swain,et al.  Color indexing , 1991, International Journal of Computer Vision.