Showing papers on "Facial recognition system published in 2001"
TL;DR: A generative appearance-based method for recognizing human faces under variation in lighting and viewpoint that 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.
Abstract: 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.
5,027 citations
TL;DR: An Automatic Face Analysis (AFA) system to analyze facial expressions based on both permanent facial features and transient facial features in a nearly frontal-view face image sequence and Multistate face and facial component models are proposed for tracking and modeling the various facial features.
Abstract: Most automatic expression analysis systems attempt to recognize a small set of prototypic expressions, such as happiness, anger, surprise, and fear. Such prototypic expressions, however, occur rather infrequently. Human emotions and intentions are more often communicated by changes in one or a few discrete facial features. In this paper, we develop an automatic face analysis (AFA) system to analyze facial expressions based on both permanent facial features (brows, eyes, mouth) and transient facial features (deepening of facial furrows) in a nearly frontal-view face image sequence. The AFA system recognizes fine-grained changes in facial expression into action units (AU) of the Facial Action Coding System (FACS), instead of a few prototypic expressions. Multistate face and facial component models are proposed for tracking and modeling the various facial features, including lips, eyes, brows, cheeks, and furrows. During tracking, detailed parametric descriptions of the facial features are extracted. With these parameters as the inputs, a group of action units (neutral expression, six upper face AU and 10 lower face AU) are recognized whether they occur alone or in combinations. The system has achieved average recognition rates of 96.4 percent (95.4 percent if neutral expressions are excluded) for upper face AU and 96.7 percent (95.6 percent with neutral expressions excluded) for lower face AU. The generalizability of the system has been tested by using independent image databases collected and FACS-coded for ground-truth by different research teams.
1,773 citations
TL;DR: However, for a task with very high dimensional data such as images, the traditional LDA algorithm encounters several difficulties, and before LDA can be used to reduce dimensionality, another procedure has to be first applied for dimensionality reduction.
1,579 citations
[...]
TL;DR: A comprehensive and critical survey of face detection algorithms, ranging from simple edge-based algorithms to composite high-level approaches utilizing advanced pattern recognition methods, is presented.
1,565 citations
TL;DR: A two-step process that allows both coarse detection and exact localization of faces is presented and an efficient implementation is described, making this approach suitable for real-time applications.
Abstract: The localization of human faces in digital images is a fundamental step in the process of face recognition. This paper presents a shape comparison approach to achieve fast, accurate face detection that is robust to changes in illumination and background. The proposed method is edge-based and works on grayscale still images. The Hausdorff distance is used as a similarity measure between a general face model and possible instances of the object within the image. The paper describes an efficient implementation, making this approach suitable for real-time applications. A two-step process that allows both coarse detection and exact localization of faces is presented. Experiments were performed on a large test set base and rated with a new validation measurement.
984 citations
TL;DR: In this article, a class-based image-based recognition and rendering with varying illumination has been proposed, based on a definition of an illumination invariant signature image which enables an analytic generation of the image space with different illumination conditions.
Abstract: The paper addresses the problem of "class-based" image-based recognition and rendering with varying illumination. The rendering problem is defined as follows: Given a single input image of an object and a sample of images with varying illumination conditions of other objects of the same general class, re-render the input image to simulate new illumination conditions. The class-based recognition problem is similarly defined: Given a single image of an object in a database of images of other objects, some of them multiply sampled under varying illumination, identify (match) any novel image of that object under varying illumination with the single image of that object in the database. We focus on Lambertian surface classes and, in particular, the class of human faces. The key result in our approach is based on a definition of an illumination invariant signature image which enables an analytic generation of the image space with varying illumination. We show that a small database of objects-in our experiments as few as two objects-is sufficient for generating the image space with varying illumination of any new object of the class from a single input image of that object. In many cases, the recognition results outperform by far conventional methods and the re-rendering is of remarkable quality considering the size of the database of example images and the mild preprocess required for making the algorithm work.
669 citations
07 Jul 2001
TL;DR: A component-based method and two global methods for face recognition and evaluate them with respect to robustness against pose changes are presented and the component system clearly outperformed both global systems on all tests.
Abstract: We present a component-based method and two global methods for face recognition and evaluate them with respect to robustness against pose changes. In the component system we first locate facial components, extract them and combine them into a single feature vector which is classified by a Support Vector Machine (SVM). The two global systems recognize faces by classifying a single feature vector consisting of the gray values of the whole face image. In the first global system we trained a single SVM classifier for each person in the database. The second system consists of sets of viewpoint-specific SVM classifiers and involves clustering during training. We performed extensive tests on a database which included faces rotated up to about 40/spl deg/ in depth. The component system clearly outperformed both global systems on all tests.
507 citations
TL;DR: An accurate and robust face recognition system was developed and tested that exploits the feature extraction capabilities of the discrete cosine transform and invokes certain normalization techniques that increase its robustness to variations in facial geometry and illumination.
Abstract: An accurate and robust face recognition system was developed and tested. This system exploits the feature extraction capabilities of the discrete cosine transform (DCT) and invokes certain normalization techniques that increase its robustness to variations in facial geometry and illumination. The method was tested on a variety of available face databases, including one collected at McGill University. The system was shown to perform very well when compared to other approaches.
496 citations
01 Jan 2001
TL;DR: Two most significant results are that changing the similarity measure produced the greatest change in performance, and that difference in performance of ±10% is needed to distinguish between algorithms.
463 citations
TL;DR: In this article, a generic modular principal component analysis (PCA) algorithm is proposed for face recognition. But the authors do not specify the design decisions of the algorithm and do not investigate the effects of these design decisions on the performance.
Abstract: Algorithms based on principal component analysis (PCA) form the basis of numerous studies in the psychological and algorithmic face-recognition literature. PCA is a statistical technique and its incorporation into a face-recognition algorithm requires numerous design decisions. We explicitly state the design decisions by introducing a generic modular PCA-algorithm. This allows us to investigate these decisions, including those not documented in the literature. We experimented with different implementations of each module, and evaluated the different implementations using the September 1996 FERET evaluation protocol (the de facto standard for evaluating face-recognition algorithms). We experimented with (i) changing the illumination normalization procedure; (ii) studying effects on algorithm performance of compressing images with JPEG and wavelet compression algorithms; (iii) varying the number of eigenvectors in the representation; and (iv) changing the similarity measure in the classification process. We performed two experiments. In the first experiment, we obtained performance results on the standard September 1996 FERET large-gallery image sets. In the second experiment, we examined the variability in algorithm performance on different sets of facial images. The study was performed on 100 randomly generated image sets (galleries) of the same size. Our two most significant results are (i) changing the similarity measure produced the greatest change in performance, and (ii) that difference in performance of +/- 10% is needed to distinguish between algorithms.
458 citations
TL;DR: The output of the PCA was submitted to a series of linear discriminant analyses which revealed three principal findings: (1) a PCA-based system can support facial expression recognition, (2) continuous two-dimensional models of emotion are reflected in the statistical structure of the Ekman and Friesen facial expressions, and (3) components for coding facial expression information are largely different to components for facial identity information.
TL;DR: By extracting uncorrelated discriminant features, face recognition could be performed with higher accuracy on lower than 16×16 resolution mosaic images and it is suggested that the optimal face image resolution can be regarded as the resolution m × n which makes the dimensionality N = mn of the original image vector space be larger and closer to the number of known-face classes.
TL;DR: The performance of the SVMs based face recognition is compared with the standard eigenface approach, and also the more recently proposed algorithm called the nearest feature line (NFL).
01 Jan 2001
TL;DR: This work develops a view-normalization approach to multi-view face and gait recognition that provides greater recognition accuracy than is obtained using the unnormalized input sequences, and that integrated face andgait recognition provides improved performance over either modality alone.
Abstract: We develop a view-normalization approach to multi-view face and gait recognition. An image-based visual hull (IBVH) is computed from a set of monocular views and used to render virtual views for tracking and recognition. We determine canonical viewpoints by examining the 3D structure, appearance (texture), and motion of the moving person. For optimal face recognition, we place virtual cameras to capture frontal face appearance; for gait recognition we place virtual cameras to capture a side-view of the person. Multiple cameras can be rendered simultaneously, and camera position is dynamically updated as the person moves through the workspace. Image sequences from each canonical view are passed to an unmodified face or gait recognition algorithm. We show that our approach provides greater recognition accuracy than is obtained using the unnormalized input sequences, and that integrated face and gait recognition provides improved performance over either modality alone. Canonical view estimation, rendering, and recognition have been efficiently implemented and can run at near real-time speeds.
TL;DR: The new coding and face recognition method, EFC, performs the best among the eigenfaces method using L(1) or L(2) distance measure, and the Mahalanobis distance classifiers using a common covariance matrix for all classes or a pooled within-class covariance Matrix.
Abstract: This paper introduces a new face coding and recognition method, the enhanced Fisher classifier (EFC), which employs the enhanced Fisher linear discriminant model (EFM) on integrated shape and texture features. Shape encodes the feature geometry of a face while texture provides a normalized shape-free image. The dimensionalities of the shape and the texture spaces are first reduced using principal component analysis, constrained by the EFM for enhanced generalization. The corresponding reduced shape and texture features are then combined through a normalization procedure to form the integrated features that are processed by the EFM for face recognition. Experimental results, using 600 face images corresponding to 200 subjects of varying illumination and facial expressions, show that (1) the integrated shape and texture features carry the most discriminating information followed in order by textures, masked images, and shape images, and (2) the new coding and face recognition method, EFC, performs the best among the eigenfaces method using L/sub 1/ or L/sub 2/ distance measure, and the Mahalanobis distance classifiers using a common covariance matrix for all classes or a pooled within-class covariance matrix. In particular, EFC achieves 98.5% recognition accuracy using only 25 features.
TL;DR: A novel approach that reformulates Fisher's discriminant ratio to a quadratic optimization problem subject to a set of inequality constraints by combining statistical pattern recognition and support vector machines is proposed.
Abstract: A novel method for enhancing the performance of elastic graph matching in frontal face authentication is proposed. The starting point is to weigh the local similarity values at the nodes of an elastic graph according to their discriminatory power. Powerful and well-established optimization techniques are used to derive the weights of the linear combination. More specifically, we propose a novel approach that reformulates Fisher's discriminant ratio to a quadratic optimization problem subject to a set of inequality constraints by combining statistical pattern recognition and support vector machines (SVM). Both linear and nonlinear SVM are then constructed to yield the optimal separating hyperplanes and the optimal polynomial decision surfaces, respectively. The method has been applied to frontal face authentication on the M2VTS database. Experimental results indicate that the performance of morphological elastic graph matching is highly improved by using the proposed weighting technique.
TL;DR: In this paper, Principal Component Analysis (PCA) is applied to a set of training similarity plots, mapping them to a lower dimensional space that contains less unwanted variation and offers better separability of the data.
Abstract: We present a novel technique for motion-based recognition of individual gaits in monocular sequences. Recent work has suggested that the image self-similarity plot of a moving person/object is a projection of its planar dynamics. Hence we expect that these plots encode much information about gait motion patterns, and that they can serve as good discriminants between gaits of different people. We propose a method for gait recognition that uses similarity plots the same way that face images are used in eigenface-based face recognition techniques. Specifically, we first apply Principal Component Analysis (PCA) to a set of training similarity plots, mapping them to a lower dimensional space that contains less unwanted variation and offers better separability of the data. Recognition of a new gait is then done via standard pattern classification of its corresponding similarity plot within this simpler space. We use the k-nearest neighbor rule and the Euclidian distance. We test this method on a data set of 40 sequences of six different walking subjects, at 30 FPS each.We use the leave-one-out cross-validation technique to obtain an unbiased estimate of the recognition rate of 93%.
TL;DR: It is shown that by decomposing a face image using wavelet transform, the low-frequency face image is less sensitive to the facial expression variations and it is proved that the spectroface representation is invariant to translation, scale and on-the-plane rotation.
01 Dec 2001
TL;DR: A method for automatically learning components by using 3-D head models, which has the advantage that no manual interaction is required for choosing and extracting components.
Abstract: We present a component-based, trainable system for detecting frontal and near-frontal views of faces in still gray images. The system consists of a two-level hierarchy of Support Vector Machine (SVM) classifiers. On the first level, component classifiers independently detect components Of a face. On the second level, a single classifier checks if the geometrical configuration of the detected components in the image matches a geometrical model of a face. We propose a method for automatically learning components by using 3-D head models, This approach has the advantage that no manual interaction is required for choosing and extracting components. Experiments show that the component-based system is significantly more robust against rotations in depth than a comparable system trained on whole face patterns.
01 Dec 2001
TL;DR: This work examines the invariance of Long-Wave InfraRed (LWIR) imagery with respect to different illumination conditions from the viewpoint of performance comparisons of two well-known face recognition algorithms applied to LWIR and visible imagery.
Abstract: A key problem for face recognition has been accurate identification under variable illumination conditions. Conventional video cameras sense reflected light so that image grayvalues are a product of both intrinsic skin reflectivity and external incident illumination, thus obfuscating the intrinsic reflectivity of skin. Thermal emission from skin, on the other hand, is an intrinsic measurement that can be isolated from external illumination. We examine the invariance of Long-Wave InfraRed (LWIR) imagery with respect to different illumination conditions from the viewpoint of performance comparisons of two well-known face recognition algorithms applied to LWIR and visible imagery. We develop rigourous data collection protocols that formalize face recognition analysis for computer vision in the thermal IR.
Patent•
05 Feb 2001TL;DR: In this paper, a mobile electronic apparatus in which biometrics information, free from being stolen or faked by an unauthorized person, is used for user verification, virtually perfectly protecting an authorized user's personal data stored in a storing section of the apparatus.
Abstract: A mobile electronic apparatus in which biometrics information, free from being stolen or faked by a unauthorized person, is used for user verification, virtually perfectly protecting an authorized user's personal data stored in a storing section of the apparatus. A verifying section compares a user's biometrics feature information with the authorized user's reference biometrics feature information to discriminate whether the fingerprint feature information for verification matches the reference fingerprint feature information. If it matches, a display control section reads out the personal data stored in the storing section and controls a display section to display the read-out personal data. The apparatus is useful when applied to a portable telephone, an electronic information terminal, or the like, so long as it has a function of user verification by biometrics information (fingerprint, palmprint, finger shape, hand shape, voiceprint, retina, iris, facial recognition, signature dynamics, blood vessel pattern, key strokes, etc.).
08 Dec 2001
TL;DR: There exists a configuration of nine point light source directions such that by taking nine images of each individual under these single sources, the resulting subspace is effective at recognition under a wide range of lighting conditions.
Abstract: Previous work has demonstrated that the image variations of many objects (human faces in particular) under variable lighting can be effectively modeled by low dimensional linear spaces. Basis images spanning this space are usually obtained in one of two ways: A large number of images of the object under different conditions is acquired, and principal component analysis (PCA) is used to estimate a subspace. Alternatively, a 3D model (perhaps reconstructed from images) is used to render virtual images under either point sources from which a subspace is derived using PCA or more recently under diffuse synthetic lighting based on spherical harmonics. In this paper we show that there exists a configuration of nine point light source directions such that by taking nine images of each individual under these single sources, the resulting subspace is effective at recognition under a wide range of lighting conditions. Since the subspace is generated directly from real images, potentially complex intermediate steps such as PCA and 3D reconstruction can be completely avoided; nor is it necessary to acquire large numbers of training images or physically construct complex diffuse (harmonic) light fields. We provide both theoretical and empirical results to explain why these linear spaces should be good for recognition.
01 Jan 2001
TL;DR: A statistical shape-fromshading model is developed to recover face shape from a single image, and to synthesize the same face under new illumination to build a simple and fast classifier that was not possible before because of a lack of training data.
Abstract: We propose a model- and exemplar-based approach for face recognition. This problem has been previously tackled using either models or exemplars, with limited success. Our idea uses models to synthesize many more exemplars, which are then used in the learning stage of a face recognition system. To demonstrate this, we develop a statistical shape-fromshading model to recover face shape from a single image, and to synthesize the same face under new illumination. We then use this to build a simple and fast classifier that was not possible before because of a lack of training data.
Patent•
10 Jul 2001TL;DR: In this article, an imaging system for providing vehicle safety features that employs face recognition software to identify and track a person is presented, which can be used for personalizing the vehicle's airbags, providing pre-crash collision avoidance, providing blind spot detection, providing vehicle crash recording, and providing a warning signal if the driver appears drowsy.
Abstract: An imaging system (50) for providing vehicle safety features that employs face recognition software to identify and track a person. The system (50) employs infrared emitters (30) that emit an infrared signal along a predetermined field-of-view, and an infrared sensor (34), such as a CMOS sensor used as a video signal array, that receives reflected infrared illumination from objects in the field-of-view. A processor (52) including the face recognition software, is employed to detect human faces to identify and track the person. Once a face is detected, it can be compared to a data base to identify the person. Various applications for the imaging system (50) for providing vehicle safety features include identifying the driver or passenger for personalizing the vehicle's airbags, providing pre-crash collision avoidance, providing blind spot detection, providing vehicle crash recording, and providing a warning signal if the driver appears drowsy.
07 Jul 2001
TL;DR: The effectiveness of the novel GFC method is shown in terms of both absolute performance indices and comparative performance against some popular face recognition schemes such as the eigenfaces method and some other Gabor wavelet based classification methods.
Abstract: This paper describes a novel Gabor feature classifier (GFC) method for face recognition. The GFC method employs an enhanced Fisher discrimination model on an augmented Gabor feature vector, which is derived from the Gabor wavelet transformation of face images. The Gabor wavelets, whose kernels are similar to the 2D receptive field profiles of the mammalian cortical simple cells, exhibit desirable characteristics of spatial locality and orientation selectivity. As a result, the Gabor transformed face images produce salient local and discriminating features that are suitable for face recognition. The feasibility of the new GFC method has been successfully tested on face recognition using 600 FERET frontal face images, which involve different illumination and varied facial expressions of 200 subjects. The effectiveness of the novel GFC method is shown in terms of both absolute performance indices and comparative performance against some popular face recognition schemes such as the eigenfaces method and some other Gabor wavelet based classification methods. In particular, the novel GFC method achieves 100% recognition accuracy using only 62 features.
28 Dec 2001
TL;DR: This paper depicts an experiment to the face recognition problem by combining eigenfaces and neural network, which can represent face pictures with several coefficients instead of having to use the whole picture.
Abstract: In this paper we depict an experiment to the face recognition problem by combining eigenfaces and neural network. Eigenfaces are applied to extract the relevant information in a face image, which are important for identification. Using this we can represent face pictures with several coefficients (about twenty) instead of having to use the whole picture. Neural networks are used to recognize the face through learning correct classification of the coefficients calculated by the eigenface algorithm. The network is first trained on the pictures from the face database, and then it is used to identify the face pictures given to it. Eight subjects (persons) were used in a database of 80 face images. A recognition accuracy of 95.6% was achieved with vertically oriented frontal views of a human face.
08 Dec 2001
TL;DR: This work extends FERET by considering when differences in recognition rates are statistically distinguishable subject to changes in test imagery and makes the source code for the algorithms, scoring procedures and Monte Carlo study available in the hopes others will extend this comparison to newer algorithms.
Abstract: The FERET evaluation compared recognition rates for different semi-automated and automated face recognition algorithms. We extend FERET by considering when differences in recognition rates are statistically distinguishable subject to changes in test imagery. Nearest Neighbor classifiers using principal component and linear discriminant subspaces are compared using different choices of distance metric. Probability distributions for algorithm recognition rates and pairwise differences in recognition rates are determined using a permutation methodology. The principal component subspace with Mahalanobis distance is the best combination; using L2 is second best. Choice of distance measure for the linear discriminant subspace matters little, and performance is always worse than the principal components classifier using either Mahalanobis or L1 distance. We make the source code for the algorithms, scoring procedures and Monte Carlo study available in the hopes others will extend this comparison to newer algorithms.
01 Jan 2001
TL;DR: A serial model for visual pattern recognition based on the primate selective attention mechanism is developed and the promise of this approach in complicated vision applications is indicated.
Abstract: A Selective Attention Based Method for Visual Pattern Recognition Albert Ali Salah (SALAH@Boun.Edu.Tr) Ethem Alpaydin (ALPAYDIN@Boun.Edu.Tr) Lale Akarun (AKARUN@Boun.Edu.Tr) Department of Computer Engineering; Bo˘gazic¸i University, 80815 Bebek Istanbul, Turkey Abstract Parallel pattern recognition requires great computational resources. It is desirable from an engineering point of view to achieve good performance with limited resources. For this purpose, we develop a serial model for visual pat- tern recognition based on the primate selective attention mechanism. The idea in selective attention is that not all parts of an image give us information. If we can attend to only the relevant parts, we can recognize the image more quickly and using less resources. We simulate the primi- tive, bottom-up attentive level of the human visual system with a saliency scheme, and the more complex, top-down, temporally sequential associative level with observable Markov models. In between, there is an artificial neu- ral network that analyses image parts and generates pos- terior probabilities as observations to the Markov model. We test our model on a well-studied handwritten numeral recognition problem, and show how various performance related factors can be manipulated. Our results indicate the promise of this approach in complicated vision appli- cations. Introduction Primates solve the problem of visual object recogni- tion and scene analysis in a serial fashion with scan- paths (Noton & Stark, 1971), which is slower but less costly than parallel recognition (Tsotsos, Culhane, Wai, Lai, Davis, Nuflo, 1995). The idea in selective attention is that not all parts of an image give us information and analysing only the relevant parts of the image in detail is sufficient for recognition and classification. The biological structure of the eye is such that a high- resolution fovea and its low-resolution periphery pro- vide data for recognition purposes. The fovea is not static, but is moved around the visual field in saccades. These sharp, directed movements of the fovea are not random. The periphery provides low-resolution informa- tion, which is processed to reveal salient points as targets for the fovea (Koch & Ullman, 1985), and those are in- spected with the fovea. The eye movements are a part of overt attention, as opposed to covert attention which is the process of moving an attentional ‘spotlight’ around the perceived image without moving the eye. In the primate brain, information from the retina is routed through the lateral geniculate nucleus (LGN) to the visual area V1 in the occipital lobe. The ‘what’ path- way, also known as the ventral pathway for anatomical reasons, goes through V4 and inferotemporal cortex (IT). The ‘where’ pathway, or the dorsal pathway, goes into the posterior parietal areas (PP) (Ungerleider & Mishkin, 1982). The ventral pathway is crucial for recognition and identification of objects, whereas the dorsal pathway me- diates the location of those objects. We should note that although recent findings point towards a distinction be- tween perception and guidance of action (Crick & Koch, 1990) instead of a distinction between different types of perception, the issue is not resolved in favour of a spe- cific theory (Milner & Goodale, 1995). The serial recognition process gathers two types of information from the image: The contents of the fovea window, and the location to which the fovea is directed. We call these ‘what’ and ‘where’ information, respec- tively (Ungerleider & Mishkin, 1982). The object is thus represented as a temporal sequence, where at each time step, the content of the fovea window and the fovea po- sition are observed. Recurrent multi-layer perceptrons were used to simul- taneously learn both the fovea features and the class sequences (Alpaydin, 1996). Other techniques are ex- plored in the literature to apply the idea of selective at- tention to classification and analysis tasks (Itti, Koch, Niebur, 1998; Rimey & Brown, 1990). Our approach is to combine a feature integration scheme (Treisman & Gelade, 1980) with a Markov model (Rimey & Brown, We use handwritten numeral recognition to test our scheme. In our database (UCI Machine Learning Repos- itory, Optdigits Database), there are ten classes (numer- als from zero to nine) with 1934 training, 946 writer- dependent cross-validation, 943 writer-dependent and 1797 writer-independent test cases. Each sample is a 32 32 binary image which is normalized to fit the bounding box. There are parallel architectures to solve this problem in the literature (Le Cun, Boser, Denker, Henderson, Howard, Hubbard, Jackel, 1989), and they have good performance, but our aim is to design a scal- able system which is applicable to problems where the input data is high-dimensional (e.g. face recognition), or not of fixed size (e.g. recognizing words in cursive hand- writing). Implementing a parallel scheme with good per- formance is not trivial in such cases. This paper is organized as follows: We first describe our model and its three levels. Then we report our simu- lation results. In the last section we summarize and indi-
Journal Article•
TL;DR: It has been over a decade since the Eigenfaces approach to automatic face recognition, and other appearancebased methods, made an impression on the computer vision research community and helped spur interest in vision systems being used to support biometrics and human-computer interface.
Abstract: SUMMARY It has been over a decade since the “Eigenfaces” approach to automatic face recognition, and other appearancebased methods, made an impression on the computer vision research community and helped spur interest in vision systems being used to support biometrics and human-computer interface. In this paper I give a personal viewof the original motivation for the work, some of the strengths and limitation of the approach, and progress in the years since. Appearance-based approaches to recognition complement feature- or shape-based approaches, and a practical face recognition system should have elements of both. Eigenfaces is not a general approach to recognition, but rather one tool out of many to be applied and evaluated in the appropriate context.