ICA FEATURES OF COLOUR AND STEREO IMAGES

Patrik O. Hoyer and Aapo Hyv#rinen

Previous work has shown that independent component analysis (ICA) applied to natural image data yields fea­ tures resembling Gabor functions and simple­cell recep­ tive øelds. This article considers the eoeects of includ­ ing chromatic and stereo information. The inclusion of colour leads to features divided into separate red/green, blue/yellow, and bright/dark channels. Stereo image data, on the other hand, leads to binocular receptive øelds which are tuned to various disparities. The sim­ ilarities between these results and observed properties of simple cells in primary visual cortex are further ev­ idence for the hypothesis that visual cortical neurons perform some type of redundancy reduction, which was one of the original motivations for ICA in the ørst place. In addition, ICA provides a principled method for feature extraction from colour and stereo images; such features could be used in image processing op­ erations such as denoising, compression, and pattern recognition.