Title :
Rotation Invariant Shape Contexts based on Feature-space Fourier Transformation
Author :
Yang, Su ; Wang, Yuanyuan
Author_Institution :
Fudan Univ., Shanghai
Abstract :
We propose a new pixel-level shape descriptor. First, shape contexts are computed. Then, 2D FFT is performed on each 2D histogram from shape contexts. Such a scheme solves the rotation-invariance problem of shape contexts based on the shift theorem of Fourier transformation while does not increase the computational complexity. Theoretical proof and experimental validation are provided.
Keywords :
Fourier transforms; computational geometry; 2D histogram; feature-space Fourier transformation; pixel-level shape descriptor; rotation invariant shape contexts; Computational complexity; Computer graphics; Computer science; Computer vision; Histograms; Pattern matching; Pattern recognition; Pixel; Robustness; Shape;
Conference_Titel :
Image and Graphics, 2007. ICIG 2007. Fourth International Conference on
Conference_Location :
Sichuan
Print_ISBN :
0-7695-2929-1
DOI :
10.1109/ICIG.2007.18
