Image distance functions for manifold learning Export

@article{citeulike:1495771, abstract = {Many natural image sets are samples of a low-dimensional manifold in the space of all possible images. When the image data set is not a linear combination of a small number of basis images, linear dimensionality reduction techniques such as PCA and ICA fail and non-linear dimensionality reduction techniques are required to automatically determine the intrinsic structure of the image set. Recent techniques such as ISOMAP and LLE provide a mapping between the images and a low-dimensional parameterization of the images. This paper specializes general manifold learning by considering a small set of image distance measures that correspond to key transformation groups observed in natural images. This results in more meaningful embeddings for a variety of applications.}, author = {Souvenir, Richard and Pless, Robert}, booktitle = {Articulated and Non-rigid motion}, citeulike-article-id = {1495771}, citeulike-linkout-0 = {http://dx.doi.org/10.1016/j.imavis.2006.01.016}, citeulike-linkout-1 = {http://linkinghub.elsevier.com/retrieve/pii/S0262885606000424}, citeulike-linkout-2 = {http://www.sciencedirect.com/science/article/B6V09-4JKHP70-1/2/82eff8e038d1000e58e01136e70c4ac9}, doi = {10.1016/j.imavis.2006.01.016}, journal = {Image and Vision Computing}, keywords = {dimensionnality-reduction, distance}, month = {March}, number = {3}, pages = {365--373}, posted-at = {2007-07-26 13:31:17}, priority = {5}, title = {Image distance functions for manifold learning}, url = {http://dx.doi.org/10.1016/j.imavis.2006.01.016}, volume = {25}, year = {2007} }

TY - JOUR ID - citeulike:1495771 L3 - citeulike-article-id:1495771 N2 - Many natural image sets are samples of a low-dimensional manifold in the space of all possible images. When the image data set is not a linear combination of a small number of basis images, linear dimensionality reduction techniques such as PCA and ICA fail and non-linear dimensionality reduction techniques are required to automatically determine the intrinsic structure of the image set. Recent techniques such as ISOMAP and LLE provide a mapping between the images and a low-dimensional parameterization of the images. This paper specializes general manifold learning by considering a small set of image distance measures that correspond to key transformation groups observed in natural images. This results in more meaningful embeddings for a variety of applications. IS - 3 JF - Image and Vision Computing EP - 373 TI - Image distance functions for manifold learning VL - 25 T2 - Articulated and Non-rigid motion SP - 365 KW - dimensionnality-reduction KW - distance AU - Souvenir, Richard AU - Pless, Robert PY - 2007/03// UR - http://dx.doi.org/10.1016/j.imavis.2006.01.016 ER -