The local low-dimensionality of natural images
This work addresses the problem of image representation for applications like denoising and compression, offering a novel approach but is incremental in advancing existing statistical models.
The authors developed a statistical model for photographic images using linear filters with locally varying Gaussian covariances, showing that images can be reconstructed nearly perfectly from local filter response covariances alone, with minimal degradation from low-rank approximations.
We develop a new statistical model for photographic images, in which the local responses of a bank of linear filters are described as jointly Gaussian, with zero mean and a covariance that varies slowly over spatial position. We optimize sets of filters so as to minimize the nuclear norms of matrices of their local activations (i.e., the sum of the singular values), thus encouraging a flexible form of sparsity that is not tied to any particular dictionary or coordinate system. Filters optimized according to this objective are oriented and bandpass, and their responses exhibit substantial local correlation. We show that images can be reconstructed nearly perfectly from estimates of the local filter response covariances alone, and with minimal degradation (either visual or MSE) from low-rank approximations of these covariances. As such, this representation holds much promise for use in applications such as denoising, compression, and texture representation, and may form a useful substrate for hierarchical decompositions.