We consider the problem of automatically recognizing human faces from frontal views with varying expression and illumination, as well as occlusion and disguise. We cast the recognition problem as one of classifying among multiple linear regression models and argue that new theory from sparse signal representation offers the key to addressing this problem. Based on a sparse representation computed by C1-minimization, we propose a general classification algorithm for (image-based) object recognition. This new framework provides new insights into two crucial issues in face recognition: feature extraction and robustness to occlusion. For feature extraction, we show that if sparsity in the recognition problem is properly harnessed, the choice of features is no longer critical. What is critical, however, is whether the number of features is sufficiently large and whether the sparse representation is correctly computed. Unconventional features such as downsampled images and random projections perform just as well as conventional features such as eigenfaces and Laplacianfaces, as long as the dimension of the feature space surpasses certain threshold, predicted by the theory of sparse representation. This framework can handle errors due to occlusion and corruption uniformly by exploiting the fact that these errors are often sparse with respect to the standard (pixel) basis. The theory of sparse representation helps predict how much occlusion the recognition algorithm can handle and how to choose the training images to maximize robustness to occlusion. We conduct extensive experiments on publicly available databases to verify the efficacy of the proposed algorithm and corroborate the above claims.

/pdf/robust-face-recognition-via-sparse-representation-4bfqv6y74o.pdf

Robust Face Recognition via Sparse Representation

As one of the most successful applications of image analysis and understanding, face recognition has recently received significant attention, especially during the past several years. At least two reasons account for this trend: the first is the wide range of commercial and law enforcement applications, and the second is the availability of feasible technologies after 30 years of research. Even though current machine recognition systems have reached a certain level of maturity, their success is limited by the conditions imposed by many real applications. For example, recognition of face images acquired in an outdoor environment with changes in illumination and/or pose remains a largely unsolved problem. In other words, current systems are still far away from the capability of the human perception system.This paper provides an up-to-date critical survey of still- and video-based face recognition research. There are two underlying motivations for us to write this survey paper: the first is to provide an up-to-date review of the existing literature, and the second is to offer some insights into the studies of machine recognition of faces. To provide a comprehensive survey, we not only categorize existing recognition techniques but also present detailed descriptions of representative methods within each category. In addition, relevant topics such as psychophysical studies, system evaluation, and issues of illumination and pose variation are covered.

Face recognition: A literature survey

/pdf/convex-analysisnoer-san-nojin-zhan-nituite-5dl2rbmoyr.pdf

Convex Analysisの二,三の進展について

We present a generative appearance-based method for recognizing human faces under variation in lighting and viewpoint. Our method exploits the fact that the set of images of an object in fixed pose, but under all possible illumination conditions, is a convex cone in the space of images. Using a small number of training images of each face taken with different lighting directions, the shape and albedo of the face can be reconstructed. In turn, this reconstruction serves as a generative model that can be used to render (or synthesize) images of the face under novel poses and illumination conditions. The pose space is then sampled and, for each pose, the corresponding illumination cone is approximated by a low-dimensional linear subspace whose basis vectors are estimated using the generative model. Our recognition algorithm assigns to a test image the identity of the closest approximated illumination cone. Test results show that the method performs almost without error, except on the most extreme lighting directions.

From few to many: illumination cone models for face recognition under variable lighting and pose

We present a new method that we call generalized discriminant analysis (GDA) to deal with nonlinear discriminant analysis using kernel function operator. The underlying theory is close to the support vector machines (SVM) insofar as the GDA method provides a mapping of the input vectors into high-dimensional feature space. In the transformed space, linear properties make it easy to extend and generalize the classical linear discriminant analysis (LDA) to nonlinear discriminant analysis. The formulation is expressed as an eigenvalue problem resolution. Using a different kernel, one can cover a wide class of nonlinearities. For both simulated data and alternate kernels, we give classification results, as well as the shape of the decision function. The results are confirmed using real data to perform seed classification.

/pdf/generalized-discriminant-analysis-using-a-kernel-approach-2p31nwoukr.pdf

Generalized Discriminant Analysis Using a Kernel Approach

A face image can be represented by a point in a feature space such as spanned by a number of eigenfaces. In methods based on nearest neighbor classification, the representational capacity of a face database depends on how prototypical face images are chosen to account for possible image variations and also how many prototypical images or their feature points are available. We propose a novel method for generalizing the representational capacity of available face database. Any two feature points of the same class (individual) are generalized by the feature line passing through the points. The feature line covers more of the face space than the feature points and thus expands the capacity of the available database. In the feature line representation, the classification is based on the distance between the feature point of the query image and each of the feature lines of the prototypical images. Experiments are presented using a data set from five databases: the MIT, Cambridge, Bern, Yale and our own. There are 620 images of 124 individuals subject to varying viewpoint, illumination, and expression. The results show that the error rate of the proposed method is about 55%-60% of that of the standard eigenface method of M.A. Turk and A.P. Pentland (1991). They also demonstrate that the recognition result can be used for inferring how the position of the input face relative to the two retrieved faces.

Generalizing capacity of face database for face recognition

Techniques that can introduce low-dimensional feature representation with enhanced discriminatory power is of paramount importance in face recognition applications. It is well known that the distribution of face images, under a perceivable variation in viewpoint, illumination or facial expression, is highly nonlinear and complex. It is therefore, not surprising that linear techniques, such as those based on principle component analysis (PCA) or linear discriminant analysis (LDA) cannot provide reliable and robust solutions to those complex face recognition problems. We propose a kernel machine based discriminant analysis method, which deals with the nonlinearity of the face patterns' distribution. The proposed method also effectively solves the "small sample size" (SSS) problem which exists in most face recognition tasks. The new algorithm has been tested, in terms of error rate performance, on the multi-view UMIST Face Database. Results indicate that the proposed methodology outperform other commonly used approaches, such as the kernel-PCA (KPCA) and the generalized discriminant analysis (GDA).

/pdf/a-kernel-machine-based-approach-for-multi-view-face-3ds3la14p1.pdf

A kernel machine based approach for multi-view face recognition

In this paper, we report the results of a comparative study on skin-color models generally used for facial region location. These include two 2D Gaussian models developed in normalized RGB and HSV color spaces respectively, a 1D lookup table model of hue histogram, and an adaptive 3D threshold box model. Also, we present a new model – called “adaptive hue lookup table”. The model is developed by introducing the so-called “Continuously Adaptive Mean Shift” (Camshift) technique into a traditional hue lookup table method. With the introduction of Camshift, the lookup table is able to adaptively adjust its parameters to fit the illumination conditions of different test images. In the experiments reported here, we compare the proposed method with the four typical skin-color filters in the scenarios of different human races and illuminations. The obtained results indicate that the proposed method reaches the best balance between false detection and detect rate.

/pdf/a-comparative-study-of-skin-color-models-fi8lg71f5b.pdf

A comparative study of skin-color models

When applied to high-dimensional pattern classification tasks such as face recognition, traditional kernel discriminant analysis methods often suffer from two problems: (1) small training sample size compared to the dimensionality of the sample (or mapped kernel feature) space, and (2) high computational complexity. In this chapter, we introduce a new kernel discriminant learning method, which attempts to deal with the two problems by using regularization and subspace de- composition techniques. The proposed method is tested by extensive experiments performed on real face databases. The obtained results indicate that the method outperforms, in terms of classification accuracy, existing kernel methods, such as kernel Principal Component Analysis and kernel Linear Discriminant Analysis, at a significantly reduced computational cost.

/pdf/kernel-discriminant-learning-with-application-to-face-2bjnbwjdby.pdf

Kernel Discriminant Learning with Application to Face Recognition

This paper addresses the issues of global optimality and training of a Feedforward Neural Network (FNN) error funtion incorporating the weight decay regularizer. A network with a single hidden-layer and a single output-unit is considered. Explicit vector and matrix canonical forms for the Jacobian and Hessian of the network are presented. Convexity analysis is then performed utilizing the known canonical structure of the Hessian. Next, global optimality characterization of the FNN error function is attempted utilizing the results of convex characterization and a convex monotonic transformation. Based on this global optimality characterization, an iterative algorithm is proposed for global FNN learning. Numerical experiments with benchmark examples show better convergence of our network learning as compared to many existing methods in the literature. The network is also shown to generalize well for a face recognition problem.

/pdf/global-feedforward-neural-network-learning-for-cwsvi3iwo2.pdf

Juwei Lu

Papers

Generalizing capacity of face database for face recognition

A kernel machine based approach for multi-view face recognition

A comparative study of skin-color models

Kernel Discriminant Learning with Application to Face Recognition

Global Feedforward Neural Network Learning for Classification and Regression