Objective methods for assessing perceptual image quality traditionally attempted to quantify the visibility of errors (differences) between a distorted image and a reference image using a variety of known properties of the human visual system. Under the assumption that human visual perception is highly adapted for extracting structural information from a scene, we introduce an alternative complementary framework for quality assessment based on the degradation of structural information. As a specific example of this concept, we develop a structural similarity index and demonstrate its promise through a set of intuitive examples, as well as comparison to both subjective ratings and state-of-the-art objective methods on a database of images compressed with JPEG and JPEG2000. A MATLAB implementation of the proposed algorithm is available online at http://www.cns.nyu.edu//spl sim/lcv/ssim/.

/pdf/image-quality-assessment-from-error-visibility-to-structural-1rlwcqe34t.pdf

Image quality assessment: from error visibility to structural similarity

Deep learning is a form of machine learning that enables computers to learn from experience and understand the world in terms of a hierarchy of concepts. Because the computer gathers knowledge from experience, there is no need for a human computer operator to formally specify all the knowledge that the computer needs. The hierarchy of concepts allows the computer to learn complicated concepts by building them out of simpler ones; a graph of these hierarchies would be many layers deep. This book introduces a broad range of topics in deep learning. The text offers mathematical and conceptual background, covering relevant concepts in linear algebra, probability theory and information theory, numerical computation, and machine learning. It describes deep learning techniques used by practitioners in industry, including deep feedforward networks, regularization, optimization algorithms, convolutional networks, sequence modeling, and practical methodology; and it surveys such applications as natural language processing, speech recognition, computer vision, online recommendation systems, bioinformatics, and videogames. Finally, the book offers research perspectives, covering such theoretical topics as linear factor models, autoencoders, representation learning, structured probabilistic models, Monte Carlo methods, the partition function, approximate inference, and deep generative models. Deep Learning can be used by undergraduate or graduate students planning careers in either industry or research, and by software engineers who want to begin using deep learning in their products or platforms. A website offers supplementary material for both readers and instructors.

Deep Learning

Training Deep Neural Networks is complicated by the fact that the distribution of each layer's inputs changes during training, as the parameters of the previous layers change. This slows down the training by requiring lower learning rates and careful parameter initialization, and makes it notoriously hard to train models with saturating nonlinearities. We refer to this phenomenon as internal covariate shift, and address the problem by normalizing layer inputs. Our method draws its strength from making normalization a part of the model architecture and performing the normalization for each training mini-batch. Batch Normalization allows us to use much higher learning rates and be less careful about initialization, and in some cases eliminates the need for Dropout. Applied to a state-of-the-art image classification model, Batch Normalization achieves the same accuracy with 14 times fewer training steps, and beats the original model by a significant margin. Using an ensemble of batch-normalized networks, we improve upon the best published result on ImageNet classification: reaching 4.82% top-5 test error, exceeding the accuracy of human raters.

/pdf/batch-normalization-accelerating-deep-network-training-by-3dkqye6twt.pdf

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

Multiresolution representations are effective for analyzing the information content of images. The properties of the operator which approximates a signal at a given resolution were studied. It is shown that the difference of information between the approximation of a signal at the resolutions 2/sup j+1/ and 2/sup j/ (where j is an integer) can be extracted by decomposing this signal on a wavelet orthonormal basis of L/sup 2/(R/sup n/), the vector space of measurable, square-integrable n-dimensional functions. In L/sup 2/(R), a wavelet orthonormal basis is a family of functions which is built by dilating and translating a unique function psi (x). This decomposition defines an orthogonal multiresolution representation called a wavelet representation. It is computed with a pyramidal algorithm based on convolutions with quadrature mirror filters. Wavelet representation lies between the spatial and Fourier domains. For images, the wavelet representation differentiates several spatial orientations. The application of this representation to data compression in image coding, texture discrimination and fractal analysis is discussed. >

/pdf/a-theory-for-multiresolution-signal-decomposition-the-530tw5pzq7.pdf

A theory for multiresolution signal decomposition: the wavelet representation

Suppose x is an unknown vector in Ropfm (a digital image or signal); we plan to measure n general linear functionals of x and then reconstruct. If x is known to be compressible by transform coding with a known transform, and we reconstruct via the nonlinear procedure defined here, the number of measurements n can be dramatically smaller than the size m. Thus, certain natural classes of images with m pixels need only n=O(m1/4log5/2(m)) nonadaptive nonpixel samples for faithful recovery, as opposed to the usual m pixel samples. More specifically, suppose x has a sparse representation in some orthonormal basis (e.g., wavelet, Fourier) or tight frame (e.g., curvelet, Gabor)-so the coefficients belong to an lscrp ball for 0<ples1. The N most important coefficients in that expansion allow reconstruction with lscr2 error O(N1/2-1p/). It is possible to design n=O(Nlog(m)) nonadaptive measurements allowing reconstruction with accuracy comparable to that attainable with direct knowledge of the N most important coefficients. Moreover, a good approximation to those N important coefficients is extracted from the n measurements by solving a linear program-Basis Pursuit in signal processing. The nonadaptive measurements have the character of "random" linear combinations of basis/frame elements. Our results use the notions of optimal recovery, of n-widths, and information-based complexity. We estimate the Gel'fand n-widths of lscrp balls in high-dimensional Euclidean space in the case 0<ples1, and give a criterion identifying near- optimal subspaces for Gel'fand n-widths. We show that "most" subspaces are near-optimal, and show that convex optimization (Basis Pursuit) is a near-optimal way to extract information derived from these near-optimal subspaces

Compressed sensing

There is no generally accepted account of the function of the second visual cortical area (V2), partly because no simple response properties robustly distinguish V2 neurons from those in primary visual cortex (V1). We constructed synthetic stimuli replicating the higher-order statistical dependencies found in natural texture images, and used them to stimulate macaque V1 and V2 neurons. Most V2 cells responded more vigorously to these textures than to control stimuli lacking naturalistic structure; V1 cells did not. fMRI measurements in humans revealed differences between V1 and V2 that paralleled the neuronal measurements. The ability of human observers to detect naturalistic structure in different types of texture was well predicted by the strength of neuronal and fMRI responses in V2 but not in V1. Together, these results reveal a novel and particular role for V2 in the representation of natural image structure.

/pdf/a-functional-and-perceptual-signature-of-the-second-visual-km8fk1rqfk.pdf

A functional and perceptual signature of the second visual area in primates

We propose the concept of quality-aware image , in which certain extracted features of the original (high-quality) image are embedded into the image data as invisible hidden messages. When a distorted version of such an image is received, users can decode the hidden messages and use them to provide an objective measure of the quality of the distorted image. To demonstrate the idea, we build a practical quality-aware image encoding, decoding and quality analysis system, which employs: 1) a novel reduced-reference image quality assessment algorithm based on a statistical model of natural images and 2) a previously developed quantization watermarking-based data hiding technique in the wavelet transform domain.

Quality-aware images

Vision and the statistics of the visual environment.

We examine a cascade encoding model for neural response in which a linear filtering stage is followed by a noisy, leaky, integrate-and-fire spike generation mechanism. This model provides a biophysically more realistic alternative to models based on Poisson (memoryless) spike generation, and can effectively reproduce a variety of spiking behaviors seen in vivo. We describe the maximum likelihood estimator for the model parameters, given only extracellular spike train responses (not intracellular voltage data). Specifically, we prove that the log-likelihood function is concave and thus has an essentially unique global maximum that can be found using gradient ascent techniques. We develop an efficient algorithm for computing the maximum likelihood solution, demonstrate the effectiveness of the resulting estimator with numerical simulations, and discuss a method of testing the model's validity using time-rescaling and density evolution techniques.

/pdf/maximum-likelihood-estimation-of-a-stochastic-integrate-and-5b3rsupknq.pdf

Maximum Likelihood Estimation of a Stochastic Integrate-and-Fire Neural Encoding Model

The visual responses of neurons in the cerebral cortex were first adequately characterized in the 1960s by D. H. Hubel and T. N. Wiesel [(1962) J. Physiol. (London) 160, 106-154; (1968) J. Physiol. (London) 195, 215-243] using qualitative analyses based on simple geometric visual targets. Over the past 30 years, it has become common to consider the properties of these neurons by attempting to make formal descriptions of these transformations they execute on the visual image. Most such models have their roots in linear-systems approaches pioneered in the retina by C. Enroth-Cugell and J. R. Robson [(1966) J. Physiol. (London) 187, 517-552], but it is clear that purely linear models of cortical neurons are inadequate. We present two related models: one designed to account for the responses of simple cells in primary visual cortex (V1) and one designed to account for the responses of pattern direction selective cells in MT (or V5), an extrastriate visual area thought to be involved in the analysis of visual motion. These models share a common structure that operates in the same way on different kinds of input, and instantiate the widely held view that computational strategies are similar throughout the cerebral cortex. Implementations of these models for Macintosh microcomputers are available and can be used to explore the models' properties.

https://www.pnas.org/content/pnas/93/2/623.full.pdf

Eero P. Simoncelli

Papers

A functional and perceptual signature of the second visual area in primates

Quality-aware images

Vision and the statistics of the visual environment.

Maximum Likelihood Estimation of a Stochastic Integrate-and-Fire Neural Encoding Model

Computational models of cortical visual processing