scispace - formally typeset
Search or ask a question

Showing papers on "Eigenface published in 2009"


Journal ArticleDOI
TL;DR: This work considers the problem of automatically recognizing human faces from frontal views with varying expression and illumination, as well as occlusion and disguise, and proposes a general classification algorithm for (image-based) object recognition based on a sparse representation computed by C1-minimization.
Abstract: 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.

9,658 citations


Book ChapterDOI
27 Jul 2009
TL;DR: This paper proposes for the first time a strongly privacy-enhanced face recognition system, which allows to efficiently hide both the biometrics and the result from the server that performs the matching operation, by using techniques from secure multiparty computation.
Abstract: Face recognition is increasingly deployed as a means to unobtrusively verify the identity of people. The widespread use of biometrics raises important privacy concerns, in particular if the biometric matching process is performed at a central or untrusted server, and calls for the implementation of Privacy-Enhancing Technologies. In this paper we propose for the first time a strongly privacy-enhanced face recognition system, which allows to efficiently hide both the biometrics and the result from the server that performs the matching operation, by using techniques from secure multiparty computation. We consider a scenario where one party provides a face image, while another party has access to a database of facial templates. Our protocol allows to jointly run the standard Eigenfaces recognition algorithm in such a way that the first party cannot learn from the execution of the protocol more than basic parameters of the database, while the second party does not learn the input image or the result of the recognition process. At the core of our protocol lies an efficient protocol for securely comparing two Pailler-encrypted numbers. We show through extensive experiments that the system can be run efficiently on conventional hardware.

546 citations


Journal ArticleDOI
01 Oct 2009
TL;DR: It is demonstrated that facial color cue can significantly improve recognition performance compared with intensity-based features and a new metric called ldquovariation ratio gainrdquo (VRG) is proposed to prove theoretically the significance of color effect on low-resolution faces within well-known subspace FR frameworks.
Abstract: In many current face-recognition (FR) applications, such as video surveillance security and content annotation in a Web environment, low-resolution faces are commonly encountered and negatively impact on reliable recognition performance. In particular, the recognition accuracy of current intensity-based FR systems can significantly drop off if the resolution of facial images is smaller than a certain level (e.g., less than 20 times 20 pixels). To cope with low-resolution faces, we demonstrate that facial color cue can significantly improve recognition performance compared with intensity-based features. The contribution of this paper is twofold. First, a new metric called ldquovariation ratio gainrdquo (VRG) is proposed to prove theoretically the significance of color effect on low-resolution faces within well-known subspace FR frameworks; VRG quantitatively characterizes how color features affect the recognition performance with respect to changes in face resolution. Second, we conduct extensive performance evaluation studies to show the effectiveness of color on low-resolution faces. In particular, more than 3000 color facial images of 341 subjects, which are collected from three standard face databases, are used to perform the comparative studies of color effect on face resolutions to be possibly confronted in real-world FR systems. The effectiveness of color on low-resolution faces has successfully been tested on three representative subspace FR methods, including the eigenfaces, the fisherfaces, and the Bayesian. Experimental results show that color features decrease the recognition error rate by at least an order of magnitude over intensity-driven features when low-resolution faces (25 times 25 pixels or less) are applied to three FR methods.

114 citations


Proceedings ArticleDOI
20 Aug 2009
TL;DR: Experimental evidence is provided which show that Polynomial and Radial Basis Function SVMs performs better than Linear SVM on the ORL Face Dataset when both are used with one against all classification.
Abstract: Automatic recognition of people has received much attention during the recent years due to its many applications in different fields such as law enforcement, security applications or video indexing. Face recognition is an important and very challenging technique to automatic people recognition. Up to date, there is no technique that provides a robust solution to all situations and different applications that face recognition may encounter. In general, we can make sure that performance of a face recognition system is determined by how to extract feature vector exactly and to classify them into a group accurately. It, therefore, is necessary for us to closely look at the feature extractor and classifier. In this paper, Principle Component Analysis (PCA) is used to play a key role in feature extractor and the SVMs are used to tackle the face recognition problem. Support Vector Machines (SVMs) have been recently proposed as a new classifier for pattern recognition. We illustrate the potential of SVMs on the Cambridge ORL Face database, which consists of 400 images of 40 individuals, containing quite a high degree of variability in expression, pose, and facial details. The SVMs that have been used included the Linear (LSVM), Polynomial (PSVM), and Radial Basis Function (RBFSVM) SVMs. We provide experimental evidence which show that Polynomial and Radial Basis Function (RBF) SVMs performs better than Linear SVM on the ORL Face Dataset when both are used with one against all classification. We also compared the SVMs based recognition with the standard eigenface approach using the Multi-Layer Perceptron (MLP) Classification criterion.

97 citations


07 Nov 2009
TL;DR: In this paper, the authors presented a methodology for face recognition based on information theory approach of coding and decoding the face image, the proposed methodology is connection of two stages - Feature extraction using principle component analysis and recognition using the feed forward back propagation Neural Network.
Abstract: Face is a complex multidimensional visual model and developing a computational model for face recognition is difficult. The paper presents a methodology for face recognition based on information theory approach of coding and decoding the face image. Proposed methodology is connection of two stages - Feature extraction using principle component analysis and recognition using the feed forward back propagation Neural Network. The algorithm has been tested on 400 images (40 classes). A recognition score for test lot is calculated by considering almost all the variants of feature extraction. The proposed methods were tested on Olivetti and Oracle Research Laboratory (ORL) face database. Test results gave a recognition rate of 97.018%.

72 citations


01 Jan 2009
TL;DR: Different face recognition approaches are referred to and primarily focuses on principal component analysis, for the analysis and the implementation is done in free software, Scilab, using SIVP toolbox for performing the image analysis.
Abstract: Face recognition systems have been grabbing high attention from commercial market point of view as well as pattern recognition field. It also stands high in researchers community. Face recognition have been fast growing, challenging and interesting area in real-time applications. A large number of face recognition algorithms have been developed from decades. The present paper refers to different face recognition approaches and primarily focuses on principal component analysis, for the analysis and the implementation is done in free software, Scilab. This face recognition system detects the faces in a picture taken by web-cam or a digital camera, and these face images are then checked with training image dataset based on descriptive features. Descriptive features are used to characterize images. Scilab's SIVP toolbox is used for performing the image analysis. Keywords—eigenfaces, PCA, face recognition, image processing, person identification, face classification, Scilab, SIVP

63 citations


Proceedings ArticleDOI
20 Jun 2009
TL;DR: The results show that in most cases it is superior to employ the average-half-face for frontal face recognition than the full face using six face recognition methods.
Abstract: We observe that the human face is inherently symmetric and we would like to exploit this symmetry in face recognition. The average-half-face has been previously shown to do just that for a set of 3D faces when using eigenfaces for recognition. We build upon that work and present a comparison of the use of the average-half-face to the use of the original full face with 6 different algorithms applied to two- and three-dimensional (2D and 3D) databases. The average-half-face is constructed from the full frontal face image in two steps; first the face image is centered and divided in half and then the two halves are averaged together (reversing the columns of one of the halves). The resulting average-half-face is then used as the input for face recognition algorithms. Previous work has shown that the accuracy of 3D face recognition using eigenfaces with the average-half-face is significantly better than using the full face. We compare the results using the average-half-face and the full face using six face recognition methods; eigenfaces, multi-linear principal components analysis (MPCA), MPCA with linear discriminant analysis (MPCALDA), Fisherfaces (LDA), independent component analysis (ICA), and support vector machines (SVM). We utilize two well-known 2D face database as well as a 3D face database for the comparison. Our results show that in most cases it is superior to employ the average-half-face for frontal face recognition. The consequences of this discovery may result in substantial savings in storage and computation time.

60 citations


Journal ArticleDOI
TL;DR: This article focuses on the strategies exploiting the temporal information, in particular those analysing: the facial motion with optical flow, the evolution of facial appearance over time with hidden Markov models (HMMs) or with various probabilistic tracking and recognition approaches, and the head motion with Gaussian mixture models.
Abstract: In this article we propose a detailed state of the art on person recognition using facial video information. We classify the existing approaches present in the scientific literature between those that neglect the temporal information, and those that exploit it even partially. Concerning the first category, we detail the extensions to video data of: eigenfaces, fisherfaces, active appearance models (AAMs), radial basis function neural networks (RBFNNs), elastic graph matching (EGM), hierarchical discriminative regression trees (HDRTs) and pairwise clustering methods. After that, we focus on the strategies exploiting the temporal information, in particular those analysing: the facial motion with optical flow, the evolution of facial appearance over time with hidden Markov models (HMMs) or with various probabilistic tracking and recognition approaches, and the head motion with Gaussian mixture models.

60 citations


Journal ArticleDOI
TL;DR: Experiments indicate that the proposed method facilitates robust face recognition under pose, illumination and expression variations and is compared with that of Eigenface, Fisherface, Subclass Discriminant Analysis, and Random Subspace LDA methods as well.
Abstract: This paper presents a novel learning approach for Face Recognition by introducing Optimal Local Basis. Optimal local bases are a set of basis derived by reinforcement learning to represent the face space locally. The reinforcement signal is designed to be correlated to the recognition accuracy. The optimal local bases are derived then by finding the most discriminant features for different parts of the face space, which represents either different individuals or different expressions, orientations, poses, illuminations, and other variants of the same individual. Therefore, unlike most of the existing approaches that solve the recognition problem by using a single basis for all individuals, our proposed method benefits from local information by incorporating different bases for its decision. We also introduce a novel classification scheme that uses reinforcement signal to build a similarity measure in a non-metric space. Experiments on AR, PIE, ORL and YALE databases indicate that the proposed method facilitates robust face recognition under pose, illumination and expression variations. The performance of our method is compared with that of Eigenface, Fisherface, Subclass Discriminant Analysis, and Random Subspace LDA methods as well.

37 citations


Proceedings ArticleDOI
01 Sep 2009
TL;DR: A general hill-climbing attack algorithm based on Bayesian adaption is used to test the vulnerability of an Eigenface-based approach for face recognition against indirect attacks, which shows a very high efficiency.
Abstract: We use a general hill-climbing attack algorithm based on Bayesian adaption to test the vulnerability of an Eigenface-based approach for face recognition against indirect attacks The attacking technique uses the scores provided by the matcher to adapt a global distribution, computed from a development set of users, to the local specificities of the client being attacked The proposed attack is evaluated on an Eigenface-based verification system using the XM2VTS database The results show a very high efficiency of the hill-climbing algorithm, which successfully bypassed the system for over 85% of the attacked accounts

30 citations


Proceedings ArticleDOI
01 Jan 2009
TL;DR: Using photometric stereo images, this paper proposes two new features for nose recognition that uses Fourier descriptors to capture the ridge shape, from the nasion to the tip, and the second uses geometric ratios, which are robustly detected using the curvature of the surface normals to locate landmarks.
Abstract: Noses are hard to conceal and relatively invariant to facial expression. Notwithstanding, their use as a biometric has been largely unexplored. Using photometric stereo images, this paper proposes two new features for nose recognition. The first of these uses Fourier descriptors to capture the ridge shape, from the nasion to the tip, and the second uses geometric ratios. Both features are robustly detected using the curvature of the surface normals to locate landmarks. Recognition results for a database of 40 individuals show that, individually, the new features out-perform an eigenface approach for an image of the nasal region. When combined they have a very respectable recognition rate for methods based on one dimensional features, indicating their potential for use within multi-feature recognition systems. (6 pages)

Proceedings ArticleDOI
01 Dec 2009
TL;DR: This paper proposes a methodology to use cylinder head models (CHMs) to track the face of a subject in multiple cameras to overcome the problem of self-occlusion by observing a person from multiple cameras with uniquely different views of the person's face and fusing the recognition results in a meaningful way.
Abstract: Face recognition from video has recently received much interest. However, several challenges for such a system exist, such as resolution, occlusion (from objects or self-occlusion), motion blur, and illumination. The aim of this paper is to overcome the problem of self-occlusion by observing a person from multiple cameras with uniquely different views of the person's face and fusing the recognition results in a meaningful way. Each camera may only capture a part of the face, such as the right or left half of the face. We propose a methodology to use cylinder head models (CHMs) to track the face of a subject in multiple cameras. The problem of face recognition from video is then transformed to a still face recognition problem which has been well studied. The recognition results are fused based on the extracted pose of the face. For instance, the recognition result from a frontal face should be weighted higher than the recognition result from a face with a yaw of 30°. Eigenfaces is used for still face recognition along with the average-half-face to reduce the effect of transformation errors. Results of tracking are further aggregated to produce 100% accuracy using video taken from two cameras in our lab.

Journal ArticleDOI
TL;DR: An innovative spectral eigenface method which uses both spatial and spectral features is proposed to improve the quality of the spectral features and to reduce the expense of the computation.
Abstract: Face recognition based on spatial features has been widely used for personal identity verification for security-related applications. Recently, near-infrared spectral reflectance properties of local facial regions have been shown to be sufficient discriminants for accurate face recognition. In this paper, we compare the performance of the spectral method with face recognition using the eigenface method on single-band images extracted from the same hyperspectral image set. We also consider methods that use multiple original and PCA-transformed bands. Lastly, an innovative spectral eigenfacemethod which uses both spatial and spectral features is proposed to improve the quality of the spectral features and to reduce the expense of the computation. The algorithms are compared using a consistent framework.

Journal ArticleDOI
TL;DR: Different from RR, LRR emphasizes on each local face region matching rather than the whole, and can not only enhance the robustness to the local variations by utilizing the spatial and geometrical information of facial components, but also avoid the dimensionality reduction in the holistic RR as a preprocessing.

Book ChapterDOI
26 May 2009
TL;DR: Experimental results show the promising aspects of new classifier when comparing with the most popular classifiers such as Nearest Neighborhood (NN), Nearest Centroid (NC), and Nearest Subspace (NS) in terms of recognition accuracy, efficiency, and numerical stability.
Abstract: In this paper, we propose a novel classification method, based on Nonnegative-Least-Square (NNLS) algorithm, for face recognition Different from traditional classifiers, in our classifier, we consider each new sample (face) as a nonnegative linear combination of training samples (faces) By forcing the nonnegative constraint on linear coefficients, we obtain the nonnegative sparse representation that automatically discriminates between those classes present in the training set Experimental results show the promising aspects of new classifier when comparing with the most popular classifiers such as Nearest Neighborhood (NN), Nearest Centroid (NC), and Nearest Subspace (NS) in terms of recognition accuracy, efficiency, and numerical stability Eigenfaces, Fisherfaces, and Laplacianfaces are performed on Yale and ORL databases as feature extraction in these experiments

Journal ArticleDOI
TL;DR: A factored covariance model is proposed for matrix data, and a method for classification using a likelihood ratio criterion is developed, which has previously been used for evaluating the strength of forensic evidence.
Abstract: A dimension reduction technique is proposed for matrix data, with applications to face recognition from images. In particular, we propose a factored covariance model for the data under study, estimate the parameters using maximum likelihood, and then carry out eigendecompositions of the estimated covariance matrix. We call the resulting method factored principal components analysis. We also develop a method for classification using a likelihood ratio criterion, which has previously been used for evaluating the strength of forensic evidence. The methodology is illustrated with applications in face recognition.

Journal ArticleDOI
TL;DR: It is shown that improvements can be made by combining gray-level profiles with Gabor wavelet features for feature extraction and hybridizing eigenface features found by Principal Component Analysis, which would provide information contained in the overall appearance of a face.
Abstract: Elastic Bunch Graph Matching is one of the well known methods proposed for face recognition. In this work, we propose several extensions to Elastic Bunch Graph Matching and its recent variant Landmark Model Matching. We used data from the FERET database for experimentations and to compare the proposed methods. We apply Particle Swarm Optimization to improve the face graph matching procedure in Elastic Bunch Graph Matching method and demonstrate its usefulness. Landmark Model Matching depends solely on Gabor wavelets for feature extraction to locate the landmarks (facial feature points). We show that improvements can be made by combining gray-level profiles with Gabor wavelet features for feature extraction. Furthermore, we achieve improved recognition rates by hybridizing Gabor wavelet with eigenface features found by Principal Component Analysis, which would provide information contained in the overall appearance of a face. We use Particle Swarm Optimization to fine tune the hybridization weights. Results of both fully automatic and partially automatic versions of all methods are presented. The best-performing method improves the recognition rate up to 22.6% and speeds up the processing time by 8 times over the Elastic Bunch Graph Matching for the fully automatic case.

Journal ArticleDOI
TL;DR: A PCA - memetic algorithm (PCA-MA) approach for feature selection for face recognition found that as far as the recognition rate is concerned, PCA- MA completely outperforms the eigenface method.
Abstract: The eigenface method that uses principal component analysis (PCA) has been the standard and popular method used in face recognition. This paper presents a PCA - memetic algorithm (PCA-MA) approach for feature selection. PCA has been extended by MAs where the former was used for feature extraction/dimensionality reduction and the latter exploited for feature selection. Simulations were performed over ORL and YaleB face databases using Euclidean norm as the classifier. It was found that as far as the recognition rate is concerned, PCA-MA completely outperforms the eigenface method. We compared the performance of PCA extended with genetic algorithm (PCA-GA) with our proposed PCA-MA method. The results also clearly established the supremacy of the PCA-MA method over the PCA-GA method. We further extended linear discriminant analysis (LDA) and kernel principal component analysis (KPCA) approaches with the MA and observed significant improvement in recognition rate with fewer features. This paper also compares the performance of PCA-MA, LDA-MA and KPCA-MA approaches.

Proceedings ArticleDOI
07 Nov 2009
TL;DR: In this paper, a real-time face detection algorithm is proposed, which uses a disparity map to estimate the face region using a calibrated stereo camera setup and then removes false positives by analyzing the disparity map.
Abstract: This paper presents a real-time face detection algorithm. It improves state-of-the-art 2D object detection techniques by additionally evaluating a disparity map, which is estimated for the face region using a calibrated stereo camera setup. First, faces are detected in the 2D images with a rapid object classifier based on haar-like features. In a second step, falsely detected faces are removed by analyzing the disparity map. In the near field of the camera, a classifier is used, which evaluates the Eigenfaces of the normalized disparity map. Thereby, the transformation into Eigenspace is learned off-line using a principal component analysis approach. In the far field, a much simpler approach determines false-positives by evaluating the relationship between the size of the face in the image and its distance to the camera. This novel combination of algorithms runs in real-time and significantly reduces the number of false-positives compared to classical 2D face detection approaches.

Journal ArticleDOI
TL;DR: The proposed system showed promising results than individual face or ear biometrics investigated in the experiments, and displayed if combined face and ear is a good technique because it offered a high accuracy and security.
Abstract: Problem statement: The study presented in this study to combined face and ear algorithms as an application of human identification. Biometric system to the detection and identification of human faces and ears developed a multimodal biometric system using eigenfaces and eigenears. Approach: The proposed system used the extracted face and ear images to develop the respective feature spaces via the PCA algorithm called eigenfaces and eigenears, respectively. The proposed system showed promising results than individual face or ear biometrics investigated in the experiments. Results: The final achieve was then used to affirm the person as genuine or an impostor. System was tested on several databases and gave an overall accuracy of 92.24% with FAR of 10% and FRR of 6.1%. Conclusion: The results display if we combined face and ear is a good technique because it offered a high accuracy and security.

Proceedings ArticleDOI
16 Dec 2009
TL;DR: This work proposes an apperence based Eigenface technique which encodes the variation among known face images which is used for recognition.
Abstract: Face recognition is one of the most active research areas in computer vision and pattern recognition with practical applications. This work proposes an apperence based Eigenface technique. PCA is used in extracting the relevant information in human faces. In this method the Eigen vectors of the set of training images are calculated which define the face space. Face images are projected on to the face space which encodes the variation among known face images. These encoded variations are used for recognition. Experiments are carried on IndianFace Database; the obtained recognition rate is 92.30%. The same training set is tested with nonface database.

Proceedings ArticleDOI
TL;DR: This paper presents a prototype system that uses facial recognition technology to monitor the authenticated user and demonstrates the feasibility of near-real-time continuous user verification for high-level security information systems.
Abstract: Information security requires a method to establish digital credentials that can reliably identify individual users. Since biometrics is concerned with the measurements of unique human physiological or behavioural characteristics, the technology has been used to verify the identity of computer or network users. Given today's heightened security requirements of military as well as other applications such as banking, health care, etc., it is becoming critical to be able to monitor the presence of the authenticated user throughout a session. This paper presents a prototype system that uses facial recognition technology to monitor the authenticated user. The objective is to ensure that the user who is using the computer is the same person that logged onto the system. A neural network-based algorithm is implemented to carry out face detection, and an eigenface method is employed to perform facial recognition. A graphical user interface (GUI) has been developed which allows the performance of face detection and facial recognition to be monitored at run time. The experimental results demonstrate the feasibility of near-real-time continuous user verification for high-level security information systems.

Proceedings ArticleDOI
Zhao Lihong1, Song Ying1, Zhu Yushi1, Zhang Cheng1, Zhang Xili1 
19 May 2009
TL;DR: By this method, when more than five face images in a face database (ORL database) are selected as training samples, with the rest as testing samples, correct recognition rate can be 97% or higher.
Abstract: Face Recognition is a very challenging topic in the field of pattern recognition, since illumination, gestures and expressions of face images are always different. In this paper, feature extraction is carried out on face images respectively through conventional methods of wavelet transform, Fourier transform, DCT, etc. Then these image transform methods are combined to process the face images. Nearest-neighbor classifiers using Euclidean distance and correlation coefficients used as similarity are adopted to recognize transformed face images. By this method, when more than five face images in a face database (ORL database) are selected as training samples, with the rest as testing samples, correct recognition rate can be 97% or higher. When five face images are from Yale face database, the correct recognition rate can be as high as 94.5%.

Journal Article
TL;DR: Experimental results on ORL, YALE and FERET face databases convince us that the proposed method provides a better representation of the class information and obtains much higher recognition accuracies.
Abstract: In this paper, an efficient local appearance feature extraction method based the multi-resolution Curvelet transform is proposed in order to further enhance the performance of the well known Linear Discriminant Analysis(LDA) method when applied to face recognition. Each face is described by a subset of band filtered images containing block-based Curvelet coefficients. These coefficients characterize the face texture and a set of simple statistical measures allows us to form compact and meaningful feature vectors. The proposed method is compared with some related feature extraction methods such as Principal component analysis (PCA), as well as Linear Discriminant Analysis LDA, and independent component Analysis (ICA). Two different muti-resolution transforms, Wavelet (DWT) and Contourlet, were also compared against the Block Based Curvelet-LDA algorithm. Experimental results on ORL, YALE and FERET face databases convince us that the proposed method provides a better representation of the class information and obtains much higher recognition accuracies. Keywords—Curvelet, Linear Discriminant Analysis (LDA) , Contourlet, Discreet Wavelet Transform, DWT, Block-based analysis, face recognition (FR).

Proceedings ArticleDOI
14 Jun 2009
TL;DR: The results achieved with the proposed method are superior in all experiments compared with the original techniques under different conditions of head pose angle, illumination and facial expression.
Abstract: One of the most successful process to accomplish human face recognition are the methods based on the Principal Component Analysis (PCA), also known as Eigenfaces. Recently, novel PCA approaches have been proposed: modular (MPCA) and two-dimensional (IMPCA). These approaches have achieved outstanding result in feature extraction and recognition. IMPCA is used for feature extraction based on 2D matrix representation and MPCA is based on image division to improve face recognition with variations like facial expressions, light and head pose. In this work we use some aspects of these methods to build a new technique called Modular Image PCA (MIMPCA). The results achieved with the proposed method are superior in all experiments compared with the original techniques under different conditions of head pose angle, illumination and facial expression.

Proceedings ArticleDOI
01 Dec 2009
TL;DR: The energy compaction property of the wavelet transform is exploited to provide a high performance face recognition system using the PCA and ZM techniques on fused and DWT approximation sub band images.
Abstract: Real time face recognition systems have become an area of growing interest due to its wide area of applications. This paper exploits the energy compaction property of the wavelet transform to provide a high performance face recognition system using the PCA and ZM techniques on fused and DWT approximation sub band images. The experimental results indicate that the wavelet sub band images enhance the face recognition accuracy by 27.2% for the PCA technique and by 3.9% for the ZM technique. Results also indicate that the wavelet approximation sub band images reduce the computational time by 50% for the PCA and by 38.46% for the ZM technique.

Journal ArticleDOI
01 Aug 2009
TL;DR: It is shown how the orientation of the robot's camera (or any active vision system) can be controlled through the estimation of the user's head pose through the initialization of the real-time tracker.
Abstract: Recently, we have proposed a real-time tracker that simultaneously tracks the 3D head pose and facial actions in monocular video sequences that can be provided by low quality cameras. This paper has two main contributions. First, we propose an automatic 3-D face pose initialization scheme for the real-time tracker by adopting a 2-D face detector and an eigenface system. Second, we use the proposed methods-the initialization and tracking-for enhancing the human-machine interaction functionality of an AIBO robot. More precisely, we show how the orientation of the robot's camera (or any active vision system) can be controlled through the estimation of the user's head pose. Applications based on head-pose imitation such as telepresence, virtual reality, and video games can directly exploit the proposed techniques. Experiments on real videos confirm the robustness and usefulness of the proposed methods.

Proceedings ArticleDOI
Lei Yunqi1, Chen Dongjie1, Yuan Meiling1, Li Qingmin1, Shi Zhen-xiang1 
28 Dec 2009
TL;DR: An approach of 3D face recognition by using of facial surface classification image and PCA is presented, which outperformed the result of using PCA method directly on the face depth image (instead of SCI) by 16.5%.
Abstract: An approach of 3D face recognition by using of facial surface classification image and PCA is presented. In the step of pre-processing, the scattered 3D points of a facial surface are normalized by surface fitting algorithm using multilevel B-splines approximation. Then, partial-ICP method is utilized to adjust 3D face model to be in the right front pose for a better recognition performance. By using the normalized facial depth image been acquired through the two previous steps, and by calculating the Gaussian and mean curvatures at each point, the surface types are classified and the classification result is used to mark different kinds of area on the facial depth image by 8 gray-levels. This achieved gray image is named as Surface Classification Image (SCI) and the SCI now represents the 3D features of the face and then it is input to the process of PCA to obtain the SCI eigenfaces to recognize the face. In the experiments conducted on 3D Facial database ZJU-3DFED of Zhejiang University, we obtained the rank-1 identification score of 94.5%, which outperformed the result of using PCA method directly on the face depth image (instead of SCI) by 16.5%.

Proceedings ArticleDOI
22 May 2009
TL;DR: An efficient face recognition algorithm based on non-negative matrix factorization (NMF) and SVM and the experimental results demonstrate the effectiveness of the proposed algorithm.
Abstract: Face recognition has received growing attention because of its wide applications. In this paper, an efficient face recognition algorithm based on non-negative matrix factorization (NMF) and SVM is proposed. The high dimension face images are first projected into a lower-dimensional subspace using NMF. The SVM classifier is then used to classify the face image into different classes. The experimental results demonstrate the effectiveness of the proposed algorithm.

Proceedings ArticleDOI
15 May 2009
TL;DR: Experimental results show the promising aspects of new approach when comparing with the most popular subspace analysis approaches in face recognition such as Eigenfaces, Fisherfaces, and Laplacianfaces in terms of recognition accuracy, efficiency, and numerical stability.
Abstract: In this paper, we present a new approach to build a more robust and efficient face recognition system. The idea is to fit the face recognition task into the new mathematical theory and algorithm of compressed sensing framework. With its beautiful theoretical results from compressed sensing, the new face recognition framework stably gives a better performance with some advantages compared to traditional approaches. Experimental results show the promising aspects of new approach when comparing with the most popular subspace analysis approaches in face recognition such as Eigenfaces, Fisherfaces, and Laplacianfaces in terms of recognition accuracy, efficiency, and numerical stability.