Title :
Shaping receptive fields for affine invariance
Author_Institution :
Massachusetts Inst. of Technol., Cambridge, MA, USA
fDate :
27 June-2 July 2004
Abstract :
The Gaussian kernel has played a central role in multi-scale methods for feature extraction and matching. In this paper, a method for shaping the filter using the local image structure is presented. We propose an optimization formulation that densely estimates the filter´s affine parameters by minimizing an objective constructed from differential feature responses and seek iterative, approximate solutions. A consequence of shaping the filters is affine invariance of the differential feature vector and it is shown that the shaped responses improve recognition performance.
Keywords :
Gaussian processes; approximation theory; feature extraction; invariance; iterative methods; optimisation; Gaussian kernel; affine invariance; approximate solutions; differential feature vector; feature extraction; filter affine parameters; multiscale methods; optimization formulation; receptive fields shaping; seek iterative; Character generation; Computer Society; Computer vision; Deformable models; Feature extraction; Gabor filters; Image recognition; Matched filters; Pattern recognition; Shape;
Conference_Titel :
Computer Vision and Pattern Recognition, 2004. CVPR 2004. Proceedings of the 2004 IEEE Computer Society Conference on
Print_ISBN :
0-7695-2158-4
DOI :
10.1109/CVPR.2004.1315236
