Refereed full papers (journals, book chapters, international conferences)

2004

@inproceedings{PaL2004,
	vgclass =	{refpap},
	author =	{Hyun-Jin Park and Te-Won Lee},
	title =	{A Hierarchical {ICA} Method for Unsupervised Learning of
	Nonlinear Dependencies in Natural Images},
	editor =	{C. G. Puntonet and A. Prieto},
	booktitle =	{Proceedings of the Fifth International Conference on
	Independent Component Analysis and Blind Signal Separation (ICA 2004)},
	address =	{Granada, Spain},
	number =	{3195},
	series =	{Lecture Notes in Computer Science},
	pages =	{1253--1261},
	publisher =	{Springer-Verlag},
	month =	{September~22--24},
	year =	{2004},
	url =	{http://www.springerlink.com/link.asp?id=bkhvhntky4gty5up},
	abstract =	{Capturing dependencies in images in an unsupervised manner
	is important for many image-processing applications and understanding
	the structure of natural image signals. Linear generative models such
	as independent component analysis (ICA) have shown to capture low level
	features such as oriented edges in images. However ICA only captures
	linear dependency due to its linear model constraints and its modeling
	capability is limited. We propose a new method for capturing nonlinear
	dependencies in natural images. It is an extension of the linear ICA
	method and builds on a hierarchical representation. It makes use of
	lower level linear ICA representation and a subsequent mixture of
	Laplacian distribution for learning the nonlinear dependencies. The
	model is learned via the EM algorithm and it can capture variance
	correlation and high order structures in a consistent manner. We
	visualize the learned variance structure and demonstrate applications
	to image segmentation and denoising.},
}