Title :
Extension of Mutual Subspace Method for Low Dimensional Feature Projection
Author :
Veljkovic, Dragana ; Robbins, Kay A. ; Rubino, Doug ; Hatsopoulos, Nicholas G.
Author_Institution :
Univ. of Texas at San Antonio, San Antonio
fDate :
Sept. 16 2007-Oct. 19 2007
Abstract :
Face recognition algorithms based on mutual subspace methods (MSM) map segmented faces to single points on a feature manifold and then apply manifold learning techniques to classify the results. This paper proposes a generic extension to MSM for analysis of features in high-throughput recordings. We apply this method to analyze short duration overlapping waves in synthetic data and multielectrode brain recordings. We compare different feature space topologies and projection techniques, including MDS, ISOMAP and Laplacian eigenmaps. Overall we find that ISOMAP shows the least sensitivity to noise and provides the best association between distance in feature space and Euclidean distance in projection space. For non-noisy data, Laplacian eigenmaps show the least sensitivity to feature space topology.
Keywords :
eigenvalues and eigenfunctions; face recognition; feature extraction; image segmentation; Euclidean distance; ISOMAP; Laplacian eigenmaps; MDS; face recognition; face segmentation; feature projection; feature space topology; image classification; manifold learning; multielectrode brain recordings; mutual subspace method; overlapping waves; projection space; Computer science; Computer vision; Energy capture; Face recognition; Image segmentation; Laplace equations; Principal component analysis; Testing; Topology; Video recording; Feature extraction; distance measurement; multidimensional systems; visualization;
Conference_Titel :
Image Processing, 2007. ICIP 2007. IEEE International Conference on
Conference_Location :
San Antonio, TX
Print_ISBN :
978-1-4244-1437-6
Electronic_ISBN :
1522-4880
DOI :
10.1109/ICIP.2007.4379189