Local Image Descriptors Using Supervised Kernel ICA

12 years 10 months ago
Local Image Descriptors Using Supervised Kernel ICA
PCA-SIFT is an extension to SIFT which aims to reduce SIFT’s high dimensionality (128 dimensions) by applying PCA to the gradient image patches. However PCA is not a discriminative representation for recognition due to its global feature nature and unsupervised algorithm. In addition, linear methods such as PCA and ICA can fail in the case of non-linearity. In this paper, we propose a new discriminative method called Supervised Kernel ICA (SKICA) that uses a non-linear kernel approach combined with Supervised ICA-based local image descriptors. Our approach blends the advantages of a low dimensionality representation, like PCA-SIFT, with supervised learning based on nonlinear properties of kernels to overcome separability limitations of nonlinear representations for recognition. Using five different test data sets we show that the SKICA descriptors produce better object recognition performance than other related approaches with the same dimensionality. The SKICA-based representation...
Masaki Yamazaki, Sidney Fels
Added 27 May 2010
Updated 27 May 2010
Type Conference
Year 2009
Authors Masaki Yamazaki, Sidney Fels
Comments (0)