DOCSLIB.ORG

Face Recognition

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Face Recognition COS 429: COMPUTER VISON Face Recognition • Intro to pattern recognition • Intro to visual recognition • PCA and Eigenfaces • LDA and Fisherfaces • generic object models for faces: the Constellation Model length length weight weight Face Recognition • Digital photography Face Recognition • Digital photography • Surveillance Face Recognition • Digital photography • Surveillance • Album organization Face Recognition • Digital photography • Surveillance • Album organization • Person tracking/id. Face Recognition • Digital photography • Surveillance • Album organization • Person tracking/id. • Emotions and expressions Face Recognition • Digital photography • Surveillance • Album organization • Person tracking/id. • Emotions and expressions • Security/warfare • Tele-conferencing • Etc. What’s ‘recognition’? vs. Identification or Discrimination What’s ‘recognition’? vs. Identification or Categorization or Discrimination Classification What’s ‘recognition’? Yes, there are faces No localization Identification or Categorization or Discrimination Classification What’s ‘recognition’? Yes, there is John Lennon No localization Identification or Categorization or Discrimination Classification What’s ‘recognition’? John Lennon Detection or Localizatoin No localization Identification or Categorization or Discrimination Classification What’s ‘recognition’? Detection or Localizatoin No localization Identification or Categorization or Discrimination Classification What’s ‘recognition’? Detection or Localizatoin No localization Identification or Categorization or Discrimination Classification What’s ‘recognition’? Segment. Detection or Localizatoin No localization Identification or Categorization or Discrimination Classification Today’s agenda 4. AdaBoost 3. Constellation model Detection or Localizatoin 1. PCA & Eigenfaces 2. LDA & Fisherfaces No localization Identification or Categorization or Discrimination Classification Eigenfaces and Fishfaces • Introduction • Techniques – Principle Component Analysis (PCA) – Linear Discriminant Analysis (LDA) • Experiments The Space of Faces = + • An image is a point in a high dimensional space – An N x M image is a point in RNM – We can define vectors in this space as we did in the 2D case [Thanks to Chuck Dyer, Steve Seitz, Nishino] Key Idea P • Images in the possible set χ = { xˆ RL } are highly correlated. • So, compress them to a low-dimensional subspace that captures key appearance characteristics of the visual DOFs. • EIGENFACES: [Turk and Pentland] USE PCA! PrincipalPrincipal ComponentComponent AnalysisAnalysis (PCA)(PCA) • PCA is used to determine the most representing features among data points. – It computes the p-dimensional subspace such that the projection of the data points onto the subspace has the largest variance among all p-dimensional subspaces. IllustrationIllustration ofof PCAPCA x2 4 x2 4 6 6 1 2 1 2 3 5 3 5 X1’ x1 x1 One projection PCA projection IllustrationIllustration ofof PCAPCA 2rd principal component x2 x1 1st principal component GoGo toto thethe handouthandout…… Mathematical Formulation Find a transformation, W, m-dimensional Orthonormal n-dimensional Total scatter matrix: Wopt corresponds to m eigen- vectors of ST Eigenfaces • PCA extracts the eigenvectors of A – Gives a set of vectors v1, v2 , v3, ... – Each one of these vectors is a direction in face space • what do these look like? Projecting onto the Eigenfaces • The eigenfaces v1, ..., vK span the space of faces – A face is converted to eigenface coordinates by AlgorithmAlgorithm Training 1. Align training images x1 , x2 , …, xN Note that each image is formulated into a long vector! 2. Compute average face u = 1/N Σ xi 3. Compute the difference image φi = xi –u AlgorithmAlgorithm 4. Compute the covariance matrix (total scatter matrix) T T ST = 1/NΣφi φi = BB , B=[φ1 , φ2 … φN ] 5. Compute the eigenvectors of the covariance matrix , W Testing 1. Projection in Eigenface Projection ωi = W (X – u), W = {eigenfaces} 2. Compare projections IllustrationIllustration ofof EigenfacesEigenfaces The visualization of eigenvectors: These are the first 4 eigenvectors from a training set of 400 images (ORL Face Database). They look like faces, hence called Eigenface. Eigenfaces look somewhat like generic faces. EigenvaluesEigenvalues ReconstructionReconstruction andand ErrorsErrors P = 4 P = 200 P = 400 Only selecting the top P eigenfaces Æ reduces the dimensionality. Fewer eigenfaces result in more information loss, and hence less discrimination between faces. SummarySummary forfor PCAPCA andand EigenfaceEigenface • Non-iterative, globally optimal solution • PCA projection is optimal for reconstruction from a low dimensional basis, but may NOT be optimal for discrimination… LinearLinear DiscriminantDiscriminant AnalysisAnalysis (LDA)(LDA) • Using Linear Discriminant Analysis (LDA) or Fisher’s Linear Discriminant (FLD) • Eigenfaces attempt to maximise the scatter of the training images in face space, while Fisherfaces attempt to maximise the between class scatter, while minimising the within class scatter. IllustrationIllustration ofof thethe ProjectionProjection Using two classes as example: x2 x2 x1 x1 Poor Projection Good Projection ComparingComparing withwith PCAPCA Variables • N Sample images: {x1,,Λ xN } • c classes: {χ1,,Λ χ c} 1 • Average of each class: μi = ∑ xk N i xk∈χ i • Total average: 1 N μ = ∑ xk N k=1 Scatters S= xμ − x− μ T • Scatter of class i: i ∑ ( k i)( k i) xk∈χ i c • Within class scatter: SSW = ∑ i i=1 c T • Between class scatter: SBχ=()∑ μi− i μi μ − () μ i=1 • Total scatter: TWBS = S + S Illustration x2 SW = S1+ S 2 S1 S2 x1 SB Mathematical Formulation (1) T After projection: yk = Wk x ~ T Between class scatter (of y’s): SWSWB = B ~ T Within class scatter (of y’s): SWSWW = W Mathematical Formulation (2) • The desired projection: ~ T SB WSWB Wargopt = max~arg= max W W T SW WSWW • How is it found ? Æ Generalized Eigenvectors SBw i= λ S iW w i i = Κ m1 , , Data dimension is much larger than the number of samples n>> N Rank(S ) ≤ N − c The matrix S W is singular: W (PCA+FLD)Fisherface T • Project with PCA to N − c space zk = Wpca k x T Wargpca = WSWmax T W T • Project with FLD to c − 1 spaceyk= W fld k z T T WWSWWpcaB pca Wargfld = max W T T WWSWWpcaW pca Illustration of FisherFace • Fisherface Results: Eigenface vs. Fisherface (1) • Input: 160 images of 16 people • Train: 159 images • Test: 1 image • Variation in Facial Expression, Eyewear, and Lighting With Without 3 Lighting 5 expressions glasses glasses conditions Eigenface vs. Fisherface (2) discussion • Removing the first three principal components results in better performance under variable lighting conditions • The Firsherface methods had error rates lower than the Eigenface method for the small datasets tested. Today’s agenda 4. AdaBoost Detection or Localizatoin 3. Constellation model 1. PCA & Eigenfaces 2. LDA & Fisherfaces No localization Identification or Categorization or Discrimination Classification Parts and Structure Literature • Fischler & Elschlager 1973 • Yuille ‘91 • Brunelli & Poggio ‘93 • Lades, v.d. Malsburg et al. ‘93 • Cootes, Lanitis, Taylor et al. ‘95 • Amit & Geman ‘95, ‘99 • et al. Perona ‘95, ‘96, ’98, ’00, ‘03 • Huttenlocher et al. ’00 • Agarwal & Roth ’02 etc… Deformations A B C D Background clutter Frontal faces Face images 3D Object recognition – Multiple mixture components 3D Orientation Tuning Orientation Tuning 100 95 90 85 80 75 % Correct % Correct 70 65 60 55 50 0 20 40 60 80 100 angle in degrees Frontal Profile Thursday this week 4. AdaBoost Detection or Localizatoin 3. Constellation model 1. PCA & Eigenfaces 2. LDA & Fisherfaces No localization Identification or Categorization or Discrimination Classification.
Load more

Recommended publications
  • A Study of the Missing Data Problems in 3D Structure Reconstruction and 2D Face Recognition
    A STUDY OF THE MISSING DATA PROBLEMS IN 3D STRUCTURE RECONSTRUCTION AND 2D FACE RECOGNITION DISSERTATION Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University By Hongjun Jia, B.E., M.E. ***** The Ohio State University 2009 Dissertation Committee: Approved by Prof. Aleix M. Martinez, Ph.D., Adviser Prof. Yuan F. Zheng, Ph.D. Adviser Prof. Alper Yilmaz, Ph.D. Graduate Program in Electrical and Computer Engineering °c Copyright by Hongjun Jia 2009 ABSTRACT Missing data problems exist in many science and engineering fields. In this thesis, we discuss two important missing data problems in computer vision and pattern recognition: 3D reconstruction with structure from motion and 2D face recognition with occlusions. The task of structure from motion for 3D reconstruction can be reduced to the problem of finding a low-rank (r) matrix that best fits an original data matrix of higher rank. The problem becomes especially difficult when the original data matrix has some missing en- tries and contains an unknown additive noise term in the remaining elements. The former problem can be solved by concatenating a set of r-column matrices which share a common, single r-dimensional solution space. Unfortunately, the number of possible submatrices is generally very large and, hence, the results obtained with one set of r-column matrices will generally be different from that captured by a different set. Ideally, we would like to find that solution which is least affected by noise. This requires that we determine which of the r-column matrices (i.e.
    [Show full text]
  • Principal Component Analysis in the Eigenface Technique for Facial Recognition
    Trinity College Trinity College Digital Repository Senior Theses and Projects Student Scholarship Spring 2012 Principal Component Analysis in the Eigenface Technique for Facial Recognition Kevin Huang Trinity College, [email protected] Follow this and additional works at: https://digitalrepository.trincoll.edu/theses Part of the Other Applied Mathematics Commons Recommended Citation Huang, Kevin, "Principal Component Analysis in the Eigenface Technique for Facial Recognition". Senior Theses, Trinity College, Hartford, CT 2012. Trinity College Digital Repository, https://digitalrepository.trincoll.edu/theses/216 Principal Component Analysis in the Eigenface Technique for Facial Recognition Kevin Huang Advisor: Professor Philip Brown Introduction Several facial recognition algorithms have been explored in the past few decades. Progress has been made towards recognition under varying lighting conditions, poses and facial expressions. In a general context, a facial recognition algorithm and its implementation can be considered as a system. The input to the facial recognition system is a two dimensional image, while the system distinguishes the input image as a user’s face from a pre-determined library of faces. Finally, the output is the discerned face image. In this project, we will examine one particular system: the Eigenface technique. Since the facial recognition problem itself deals with images, each pixel in an image is considered as a coordinate in an n-dimensional space, where n is the total number of pixels per image. Now suppose that there is a collection of several images that forms a library of faces that serve as candidates for recognition. Since each image is a point in n-space, and n can be a relatively large number, it would be convenient and computationally efficient to reduce the dimensionality of the data set.
    [Show full text]
  • Sparse Shape Modelling for 3D Face Analysis
    Sparse Shape Modelling for 3D Face Analysis Stephen Clement Doctor of Philosophy University of York Computer Science September 2014 Abstract This thesis describes a new method for localising anthropometric landmark points on 3D face scans. The points are localised by fitting a sparse shape model to a set of candidate landmarks. The candidates are found using a feature detector that is designed using a data driven methodology, this approach also informs the choice of landmarks for the shape model. The fitting procedure is developed to be robust to missing landmark data and spurious candidates. The feature detector and landmark choice is determined by the performance of different local surface descriptions on the face. A number of criteria are defined for a good landmark point and good feature detector. These inform a framework for measuring the performance of various surface descriptions and the choice of parameter values in the surface description generation. Two types of surface description are tested: curvature and spin images. These descriptions, in many ways, represent many aspects of the two most common approaches to local surface description. Using the data driven design process for surface description and landmark choice, a feature detector is developed using spin images. As spin images are a rich surface description, we are able to perform detection and candidate landmark labelling in a single step. A feature detector is developed based on linear discriminant analysis (LDA). This is compared to a simpler detector used in the landmark and surface description selection process. A sparse shape model is constructed using ground truth landmark data.
    [Show full text]
  • Beyond Eigenfaces: Probabilistic Matching for Face Recognition
    M.I.T Media Lab oratory Perceptual Computing Section Technical Rep ort No. 443 App ears in: The 3rd IEEE Int'l ConferenceonAutomatic Face & GestureRecognition, Nara, Japan, April 1998 Beyond Eigenfaces: Probabilistic Matching for Face Recognition Baback Moghaddam, Wasiuddin Wahid and Alex Pentland MIT Media Lab oratory, 20 Ames St., Cambridge, MA 02139, USA. Email: fbaback,wasi,[email protected] facial expressions of the same individual ) and extrapersonal Abstract variations (corresp onding to variations b etween di er- E We prop ose a novel technique for direct visual ent individu al s). Our similarity measure is then expressed matching of images for the purp oses of face in terms of the probabili ty recognition and database search. Sp eci call y, we argue in favor of a probabilistic measure of S (I ;I )=P ( 2 )=P ( j) (1) 1 2 I I similarity, in contrast to simpler metho ds which where P ( j) is the aposteriori probabilitygiven by I are based on standard L norms (e.g., template 2 Bayes rule, using estimates of the likeliho o d s P (j )and I matching) or subspace-restricted norms (e.g., P (j )which are derived from training data using an E eigenspace matching). The prop osed similarity ecient subspace metho d for density estimation of high- measure is based on a Bayesian analysis of image dimensional data [6]. This Bayesian (MAP) approach di erences: we mo del twomutually exclusive can also b e viewed as a generalized nonlinear extension of classes of variation b etween two facial images: Linear Discriminant Analysis (LDA) [8, 3] or \FisherFace" intra-personal (variations in app earance of the techniques [1] for face recognition.
    [Show full text]
  • Robust Face Recognition by Combining Projection-Based Image Correction and Decomposed Eigenface
    View metadata, citation and similar papers at core.ac.uk brought to you by CORE provided by Okayama University Scientific Achievement Repository Engineering Industrial & Management Engineering fields Okayama University Year 2004 Robust face recognition by combining projection-based image correction and decomposed eigenface Takeshi Shakunaga Fumihiko Sakaue Okayama University Okayama University Kazuma Shigenari Okayama University This paper is posted at eScholarship@OUDIR : Okayama University Digital Information Repository. http://escholarship.lib.okayama-u.ac.jp/industrial engineering/24 Robust Face Recognition by Combining Projection-Based Image Correction and Decomposed Eigenface Takeshi Shakunaga Fumihiko Sakaue Kazuma Shigenari Department of Information Technology, Faculty of Engineering Okayama University Okayama-shi, Okayama 700-8530, Japan E-mail: {shaku,sakaue}@chino.it.okayama-u.ac.jp Abstract lected in the registration stage, the EF can be constructed by Principal Component Analysis (PCA). However, an EF for This paper presents a robust face recognition method which an individual cannot be stably constructed when an insuf- can work even when an insufficient number of images are ficient number of sample images are available or when the registered for each person. The method is composed of im- sample images have been taken under very similar light- age correction and image decomposition, both of which are ing conditions. In these situations, realizing illumination- specified in the normalized image space (NIS). The image insensitive identification requires some refinements of the correction[1, 2] is realized by iterative projections of an eigenface method. In our previous paper [3], we analyzed image to an eigenspace in NIS. It works well for natural the eigenface approach and proposed concepts of virtual images having various kinds of noise, including shadows, eigenfaces and the decomposed eigenface for the forego- reflections, and occlusions.
    [Show full text]
  • 10-701 Machine Learning, Fall 2011: Homework 5
    10-701 Machine Learning, Fall 2011: Homework 5 Due 12/5 at the beginning of class. November 17, 2011 1 Principal Component Analysis: Eigenfaces [25 points, Bin] PCA is applied to recognize faces by Matthew Turk and Alex Pentland in their \Eigenface" ap- proach, which is considered as the first successful example of face recognition. In this question, you will implement your own face recognition system. Please print out your codes and attach them at the end of your answer sheet for this question. 1.1 ORL Face Data The face data set used in this question is the ORL database of faces1. There are 4 files in HW4 data.zip: • X train.csv: each of the 360 lines contains data for a 112 × 92 face image. If you are using MATLAB, you can load the images using 'data = csvread('X train.csv')', and display the i-th image using 'colormap(gray); imagesc(reshape(data(i,:),112,92))'. • Y train.csv: each of the 360 lines contains the label for the corresponding line in X train.csv • X test.csv: 40 test images organized in the same way as in X train.csv • Y test.csv: labels for face images in X test.csv 1.2 Calculating Eigenfaces This step will only use X train.csv: • [3 points] First compute the mean Ψ of the 360 training faces, and subtract the mean from each training data point. • [7 points] Compute the covariance matrix for the 360 training faces, then apply eigen- decomposition on the obtained covariance matrix. In order to save computation time, you could use 'eigs' in MATLAB to compute only the eigenvectors corresponding to the 50 largest eigenvalues.
    [Show full text]
  • Face Recognition Using Principle Component Analysis, Eigenface and Neural Network
    SENSORS, SIGNALS, VISUALIZATION, IMAGING, SIMULATION AND MATERIALS Face Recognition using Principle Component Analysis, Eigenface and Neural Network Mayank Agarwal Nikunj Jain Student Member IEEE Student Jaypee Institute of Information Technology Jaypee Institute of Information Technology University University Noida ,India Noida ,India [email protected] [email protected] Mr. Manish Kumar Himanshu Agrawal Sr. Lecturer (ECE) Student Member IEEE Jaypee Institute of Information Technology Jaypee Institute of Information Technology University University Noida, India Noida, India [email protected] [email protected] Abstract-Face is a complex multidimensional visual Face recognition has become an important issue in many model and developing a computational model for face applications such as security systems, credit card recognition is difficult. The paper presents a methodology verification, criminal identification etc. Even the ability to for face recognition based on information theory approach merely detect faces, as opposed to recognizing them, can of coding and decoding the face image. Proposed be important. methodology is connection of two stages – Feature Although it is clear that people are good at face extraction using principle component analysis and recognition, it is not at all obvious how faces are encoded recognition using the feed forward back propagation or decoded by a human brain. Human face recognition Neural Network. has been studied for more than twenty years. Developing The algorithm has been tested on 400 images (40 classes). a computational model of face recognition is quite A recognition score for test lot is calculated by difficult, because faces are complex, multi-dimensional considering almost all the variants of feature extraction.
    [Show full text]
  • Towards Face Recognition Using Eigenface
    (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 7, No. 5, 2016 Towards Face Recognition Using Eigenface Md. Al-Amin Bhuiyan Department of Computer Engineering King Faisal University Hofuf, Al-Ahsa 31982, Saudi Arabia Abstract—This paper presents a face recognition system Liu and Wechsler [6] established a Gabor filter based employing eigenface-based approach. The principal objective of identification method for dimensional reduction employing this research is to extract feature vectors from images and to Fisher linear discriminate model. Kirby [8] and Terzopoulos, reduce the dimension of information. The method is implemented et al. [9] analysed the facial images on the bsis of facial on frontal view facial images of persons to explore a two- characteristics. Wu [10] and Manjunath [11], et al. performed dimensional representation of facial images. The system is face recognition with feature vectors extracted from profile organized with RMS (Root Mean Square) contrast scaling silhouettes. Kerin, et al. [12] have prposed a face recognition technique employed for pre-processing the images to adjust with system employing neural network in tendem with self- poor lighting conditions. Experiments have been conducted using organizing feature map. Nakamura, et al. [13] utilized Carnegie Mellon University database of human faces and isodensity maps to establish a face recognition system. Yullie, University of Essex Computer Vision Research Projects dataset. et al. [14] extracted feature vectors from the facial Experimental results indicate that the proposed eigenface-based components like eyes, nose, mouth, and and employed approach can classify the faces with accuracy more than 80% in all cases. deformable templates to the extraction of contours for facial images and outlined a face recognition procedure.
    [Show full text]
  • Performance Evaluation of PCA and LDA for Face Recognition S
    ISSN: 2278 – 909X International Journal of Advanced Research in Electronics and Communication Engineering (IJARECE) Volume 2, Issue 2, February 2013 Performance Evaluation of PCA and LDA for Face Recognition S. K. Hese, M. R. Banwaskar Department of Electronics & Telecommunication, MGM’s College of Engineering Nanded Near Airport, Nanded, Maharashtra, India Abstract — Automated face recognition has become a major [3], [4]. The Fisherfaces is an enhancement of the Eigenface field of interest. In this field several facial recognition algorithms method. The Eigenface method uses PCA for dimensionality have been explored in the past few decades. Progress has been reduction, thus, yields projection directions that maximize the made towards recognition under varying lighting conditions, total scatter across all classes, i.e. across all images of all poses and facial expressions. In a general context, a facial faces. The PCA projections are optimal for representation in a recognition algorithm and its implementation can be considered as a system. The input to the facial recognition system is a two low dimensional basis, but they may not be optional from a dimensional image, while the system distinguishes the input discrimination standpoint. Instead, the Fisherfaces method image as a user’s face from a pre-determined library of faces. uses Fisher’s Linear Discriminant Analysis (FLDA or LDA) Finally, the output is a discerned face image. which maximizes the ratio of between-class scatter to that of within-class scatter [5]. This paper discusses different appearance based face recognition techniques. The experimentation includes the use of At one level, PCA and LDA are very different: LDA image preprocessing techniques followed by most popular is a supervised learning technique that relies on class labels, dimensionality reduction algorithms based on PCA and LDA.
    [Show full text]
  • SYMMETRIC EIGENFACES 1. Introduction. Facial Recognition Is A
    SYMMETRIC EIGENFACES MILI I. SHAH ∗ Abstract. Over the years, mathematicians and computer scientists have produced an extensive body of work in the area of facial analysis. Several facial analysis algorithms have been based on mathematical concepts such as the singular value decomposition (SVD). The SVD is generalized in this paper to take advantage of the mirror symmetry that is inherent in faces, thereby developing a new facial recognition algorithm: the symmetry preserving singular value decomposition (SPSVD). The SPSVD recognizes faces using half the total computations employed by the conventional SVD. Moreover, the SPSVD provides more accurate recognition, even in the presence of noise in the form of light variation and/or facial occlusion. Key words. Facial analysis, eigenfaces, reflection, symmetry, SVD, PCA, SPSVD AMS subject classifications. 15A18, 65F15 1. Introduction. Facial recognition is a technology that is fast becoming criti- cal for contemporary security applications. For instance, many airports employ facial recognition algorithms in their closed-circuit televisions to identify - in real-time - potential terrorists. Future applications of facial recognition may include the use of these algorithms in facilitating user identification by, e.g., automated teller machines (ATMs). Many facial recognition algorithms are based on the singular value decomposition (SVD) [3,4,7]. The SVD takes an image of a given face and compares it to facial images stored in a database, using a reduced representation of the face. The representation in the database that most closely matches the face of the person being scanned is returned, as is a measure of difference between the two faces. Thus, this approach can provide an estimate of the probability that a match has been detected.
    [Show full text]
  • Eigenface-Based Facial Recognition
    Eigenface-based facial recognition Dimitri PISSARENKO December 1, 2002 1 General This document is based upon Turk and Pentland (1991b), Turk and Pentland (1991a) and Smith (2002). 2 How does it work? The task of facial recogniton is discriminating input signals (image data) into several classes (persons). The input signals are highly noisy (e.g. the noise is caused by differ- ing lighting conditions, pose etc.), yet the input images are not completely random and in spite of their differences there are patterns which occur in any input signal. Such patterns, which can be observed in all signals could be – in the domain of facial recog- nition – the presence of some objects (eyes, nose, mouth) in any face as well as relative distances between these objects. These characteristic features are called eigenfaces in the facial recognition domain (or principal components generally). They can be ex- tracted out of original image data by means of a mathematical tool called Principal Component Analysis (PCA). By means of PCA one can transform each original image of the training set into a cor- responding eigenface. An important feature of PCA is that one can reconstruct recon- struct any original image from the training set by combining the eigenfaces. Remember that eigenfaces are nothing less than characteristic features of the faces. Therefore one could say that the original face image can be reconstructed from eigenfaces if one adds up all the eigenfaces (features) in the right proportion. Each eigenface represents only certain features of the face, which may or may not be present in the original im- age.
    [Show full text]
  • Face Recognition Using Multiple Eigenface Subspaces
    Journal of Engineering and Technology Research Vol. 2(8), pp. 139-143, August 2010 Available online at http://www.academicjournals.org/JETR ISSN 2006-9790 ©2010 Academic Journals Full Length Research Paper Face recognition using multiple eigenface subspaces P. Aishwarya1* and Karnan Marcus2 1Department of Computer Science and Engineering, Atria Institute of Technology, Bangalore – 24, India. 2Department of Computer Science and Engineering, Tamil Nadu College of Engineering, Coimbatore, India. Accepted 14 June, 2010 Face recognition is grabbing more attention in the area of network information access. Areas such as network security and content retrieval benefit from face recognition technology. In the proposed method, multiple face eigensubspaces are created, with each one corresponding to one known subject privately, rather than all individuals sharing one universal subspace as in the traditional eigenface method. Compared with the traditional single subspace face representation, the proposed method captures the extra personal difference to the most possible extent, which is crucial to distinguish between individuals, and on the other hand, it throws away the most intrapersonal difference and noise in the input. Our experiments strongly support the proposed idea, in which 20% improvement of performance over the traditional “eigenface” has been observed when tested on the same face base. Key words: Face recognition, eigenspace, subspaces. INTRODUCTION Face recognition (FR) has emerged as one of the most In this paper, the method proposed is derived from the extensively studied research topics that spans multiple traditional Eigenface but differs from it in essence. In disciplines such as pattern recognition, signal processing Eigenface, each face image is represented as a point in a and computer vision.
    [Show full text]