A Simple Feature Extraction for High Dimensional Image Representations
We investigate a method to find local clusters in low dimensional subspaces of high dimensional data, e.g. in high dimensional image descriptions. Using cluster centers instead of the full set of data will speed up the performance of learning algorithms for object recognition, and will possibly also improve performance because overfitting might be avoided. For the artificial data used in this work, our method outperforms the current standard method for feature extraction from high dimensional data.