Discriminative Hessian Eigenmaps for face recognition
|
|
- Patrick Elijah Gilbert
- 5 years ago
- Views:
Transcription
1 tle Dscrmnatve essan Egenmaps or ace recognton Author(s) S S; ao D; Chan KP Ctaton he 200 IEEE Internatonal Conerence on Acoustcs Speech and Sgnal Processng (ICASSP) Dallas. 4-9 March 200. In IEEE Internatonal Conerence on Acoustcs Speech and Sgnal Processng Proceedngs 200 p Issued Date 200 UR Rghts IEEE Internatonal Conerence on Acoustcs Speech and Sgnal Processng Proceedngs. Copyrght IEEE.; 200 IEEE. Personal use o ths materal s permtted. owever permsson to reprnt/republsh ths materal or advertsng or promotonal purposes or or creatng new collectve works or resale or redstrbuton to servers or lsts or to reuse any copyrghted component o ths work n other works must be obtaned rom the IEEE.; hs work s lcensed under a Creatve Commons Attrbuton-NonCommercal-NoDervatves 4.0 Internatonal cense.
2 DISCRIMINAIVE ESSIAN EIGENMAPS FOR FACE RECOGNIION S S Dacheng ao 2 Kwok-Png Chan Department o Computer Scence Unversty o ong Kong ong Kong 2 School o Computer Engneerng Nanyang echnologcal Unversty Sngapore ABSRAC Dmenson reducton algorthms have attracted a lot o attentons n ace recognton because they can select a subset o eectve and ecent dscrmnatve eatures n the ace mages. Most o dmenson reducton algorthms can not well model both the ntra-class geometry and nterclass dscrmnaton smultaneously. In ths paper we ntroduce the Dscrmnatve essan Egenmaps () a novel dmenson reducton algorthm to address ths problem. wll consder encodng the geometrc and dscrmnatve normaton n a local patch by mproved essan Egenmaps and margn mamzaton respectvely. Emprcal studes on publc ace database thoroughly demonstrate that s superor to popular algorthms or dmenson reducton e.g. FDA PP and DA. Inde erms Dmenson Reducton Manold earnng Face Recognton.. INRODUCION Dmenson reducton [3][] plays an mportant role n varous tasks n computer vson e.g. ace recognton. A key role or ace recognton s the dstance or smlarty between ace mages whch can be solved va dmenson reducton as dmenson reducton perorms the recognton by enlargng the smlarty among the ntra-class samples and mamzng the derence among the nter-class samples n a subspace rather than the orgnal eature space. A dmenson reducton algorthm projects the orgnal hgh-dmensonal eature space to a low-dmensonal subspace where specc statstcal propertes can be well preserved. For eample prncple component analyss (PCA) [] one o the most popular unsupervsed dmenson reducton algorthms mamzes the varance o the data n the projected subspace; Fsher s lnear dscrmnatve analyss (FDA) [2] the most tradtonal supervsed dmenson reducton algorthm mnmzes the trace rato between the wthn class scatter and the between class scatter so that the Gaussan dstrbuted samples can be well separated n the selected subspace; localty preservng projectons (PP) [4] preserves the local geometry o samples by processng an undrected weghted graph that represents the neghbourhood relatons o parwse samples; Margnal Fsher analyss () [2] consders both the ntra-class geometry and nteracton o samples rom derent classes; Dscrmnatve localty algnment (DA) [5] preserves the dscrmnatve normaton by mamzng the dstance among the nter-class samples and mnmzng the dstance among the ntra-class samples over the local patch o each sample. owever the geometrc and dscrmnatve normaton n these dmenson reducton algorthms are not well modeled e.g. DA does not consder the geometrc normaton; gnores the dscrmnatve normaton o non-margnal samples rom derent classes. By usng the patch algnment ramework [6] we can model both the ntra-class local geometry and the nter-class dscrmnatve normaton convenently. In partcular or each sample and ts assocated patch (neghbours o the sample) t s mportant to consder the ollowng two propertes: ) the ntra-class local geometry can be represented by the local tangent space whch s locally sometrc to the manold o the ntra-class nearest samples o the patch; and 2) the nter-class dscrmnatve normaton can be represented by the margn between the ntra-class neghbor samples and the nter-class nearest samples o the patch. Because the method used or local geometry representaton s smlar to essan Egenmaps [7] the proposed dmenson reducton algorthm s termed the Dscrmnatve essan Egenmaps or or short. he rest o ths paper s organzed as ollows. Secton 2 ntroduces the proposed Dscrmnatve essan Egenmaps (). Secton 3 shows the results o thoroughly emprcal studes. Secton 4 concludes. 2. DISCRIMINAIVE ESSIAN EIGENMAPS hs Secton presents the dscrmnatve essan Egenmaps or or short to solve the ace recognton tasks. In D d we try to nd an optmal lnear mappng W R D so that t can project R to a low-dmensonal space as d y W R. In ths learned low-dmensonal space characterzes two specc propertes: /0/$ IEEE 5586 ICASSP 200
3 . he local geometry property - nearby samples n the orgnal Eucldean space are close to each other n the learned subspace. 2. he dscrmnatve property - samples rom derent classes can be well separated n the learned subspace. In summary the dscrmnatve normaton as well as the local geometry wll be well modeled n the. 2.. Moded essan Egenmaps Emprcally ntra-class geometry s useul or classcaton. essan Egenmaps [7] s a geometry preservaton manold learnng method that can recover the underlyng parameterzaton o a manold M embedded n a hghdmensonal space the manold M s locally sometrc to d an open and connected subset o R. Because the parameter space need not be conve n essan Egenmaps t can be appled to model a nonconve manold e.g. an S- curve surace wth a hole. hereore we adapt essan Egenmaps n to preserve the local geometry or dmenson reducton. essan Egnmaps nds the (d+)-dmensonal nullspace o where s the essan matr o a smooth mappng.e. : M R can be. hs calculated by usng 2 d wheren s the essan o on the patch k and the correspondng output n low-dmensonal space s Y y y y k. he M a Eucldean space tangental to M at tangent plane s an orthogonal coordnate system. In order to estmate we calculate the local coordnate system o and each sample n on the tangent plane M has ts own local coordnate can be estmated by usng. owever essan Egenmaps cannot be appled to many practcal applcatons e.g. ace recognton because t requres that k d where k s the number o the neghbourng samples and d s the dmenson o the subspace. It s dcult to guarantee ths condton because M. Aterwards ths we have a lmted number o samples. We propose to overcome ths problem by perormng PCA on M at and orthnormalzng the d-dmensonal representaton to obtan the tangent coordnate n F M. he ollowng steps or the moded essan Egenmaps are smlar to those n essan Egenmaps. Under the patch algnment ramework the objectve uncton or the moded essan Egenmaps to preserve the local geometry on a local patch Y can be wrtten as where normaton o the patch and y tr Y Y tr Y Y () encodes the local geometry y s the local geometry representaton. Under the help o local geometrc normaton can be urther preserved Margn Mamzaton As or classcaton however t s nsucent to only retan the local geometry because no labelng normaton s taken nto account. o urther eplot the dscrmnatve power lke the denton o the local geometry we can dene a new margn mamzaton [3] based scheme or dscrmnatve normaton preservaton over patches. In partcular or each sample assocated wth a patch k M k wheren 2 k.e. the k nearest samples o are rom the same class as and.e. the other k 2 nearest samples o are rom derent classes aganst we dene the margn as the average derence between two knds o dstances on ths patch. One s called nter-class dstance that s the dstance between and samples takng derent labels.e. ; the other s called ntra-class dstance that s the dstance between and samples sharng the same label.e. k. Bascally n the patch M s lowdmensonal representaton Y y y y k y M y k 2 we epect the margn between ntra-class and nter-class samples wll be mamzed as large as possble.e. 2 k 2 y y y j p y p k. (2) 2 j k On the other hand based on (2) we try to mnmze the bellowng objectve uncton: k 2 2 M y y y j y yp j k p e k tr Y dag k M w e k I 2 kk Y 2 M I (3) k k 2 tr Y Y where M M M k w / k.../ k -/k /k 2 ; Ik s the 5587
4 2 k k k k dentty matr ;... k k e R 2 2 k M normaton representaton. ; k w w j j and w dag w M y s the margn 2.3. Dscrmnatve essan Egenmaps () By usng the results obtaned rom the prevous subsectons we can obtan the optmzaton ramework to learn the projecton matr W whch can utlze both the local geometry and the dscrmnatve normaton. Because the margn representaton M y and the local geometry representaton y are dened over patches and each patch has ts own coordnate system algnment strategy s adopted here to buld a global coordnate or all patches dened or the tranng samples. As a consequence the objectve uncton or to solve the dmenson reducton problem s gven by l W arg mn M y y (4) Dd WR where s the tunng parameter. I we dene two selecton matres S and S M whch select samples n the th patch rom all the tranng samples Y y y2 yl or constructng M y and y respectvely. hereore Y Y S and Y Y M SM wth Y representng the patch or the local geometry preservaton and Y M denotng the patch or margn mamzaton. Ater pluggng () and (3) the objectve uncton n (4) wll turn to l tr Y Y M M M W arg mn 5 Dd WR tr Y Y tr YS M l M YS M arg mn Dd WR tr YS YS l SM MS M arg mn tr Y Y Dd WR S S arg mn tr YY Dd WR l where S S M M MS S s the algnment matr encodng both the local geometry and the dscrmnatve normaton. For lnearzaton Y W s usually consdered where W s the projecton matr. We can mpose derent constrants e.g. Y Y I or W W I to unquely determne Y. he constrant W W I wll be adopted throughout the paper. Under ths constrant and Y W the soluton o (5) can be obtaned by usng the conventonal agrangan multpler method [0] or the generalzed egenvalue decomposton [8]. 3. EPERIMENS In ths Secton we justy our proposed algorthm wth our representatve dmenson reducton algorthms whch are the Fsher s lnear dscrmnant analyss (FDA) [2] the localty preservaton projectons (PP) [4] wth the supervsed settng the margnal Fsher s analyss () [2] and dscrmnatve localty algnment (DA) [6] or ace recognton based on a publc database: CMU-PIE dataset [9]. Fgure. Sample mages rom CMU-PIE database he CMU-PIE dataset contans 4368 mages o 68 people under 3 derent poses 43 derent llumnaton condtons and 4 derent epressons and we randomly select 0 mages per ndvdual n the CMU-PIE dataset n ths eperment. Eample ace mages rom the CMU-PIE database are shown n Fgure. he mages rom CMU-PIE used or our eperments are o sze 3232 n raw pel. In the tranng stage we learn the projecton matr W rom each nvolved algorthm on the tranng samples. In the testng stage each testng sample wll be projected nto the low-dmensonal space by W and ater that nearest-neghbor rule (NN) s appled to predct label o the test mage n the selected subspace. We randomly select p (= 4 5 6) mages per ndvdual or tranng n the database and use the remanng mages or testng. All trals are repeated ten tmes and then the average recognton rates are calculated. Fgure 2 shows the results o aganst FDA PP and DA wth regard to ace recognton accuracy under derent dmensons. able provdes the best recognton rate or each algorthm. It also provdes the optmal values o k k 2 and or whch are tuned by the cross valdaton. 5588
5 00 4 ran 00 5 ran 00 6 ran Recognton Rate (%) FDA PP DA Subspace dmensons Recognton Rate (%) FDA PP DA Subspace dmensons FDA PP DA Subspace dmensons Fgure 2. Recognton rate vs. dmenson reducton on the CMU-PIE database under derent splts. Recognton Rate (%) able. Best recognton rates (%) on CMU-PIE database. he numbers n the parentheses are the subspace dmensons. For he numbers n the parentheses rom let to rght are the subspace dmensons k and respectvely. FDA PP DA 4 ran 8.79(67) 82.33(68) 88.58(78) 86.(39).86(2936) 5 ran 88.94(67) 89.38(67).29().94(62) 94.44(47365) 6 ran 92.58(69) 93.7(67) 92.67() 93.46(34) 96.8(473) As shown n Fgure 2 and able outperorms conventonal algorthms or at least can obtan a comparable perormance n comparng wth the conventonal algorthms because can precsely model both the ntra-class geometry and the nter-class dscrmnatve normaton n the local patch. 4. CONCUSION In ths paper we have proposed a novel lnear dmenson reducton algorthm termed Dscrmnatve essan Egenmaps (). s superor to the conventonal dmensonalty reducton algorthms because t ocuses on accurately modelng both the ntra-class geometry and nterclass dscrmnaton n the local patch. Emprcal studes on ace recognton tasks demonstrate that s more eectve than conventonal algorthms. 5. ACKNOWEDGEMEN S. S and K.-P. Chan thank the support rom KU-SPF Grant (project number 0006). D. ao thanks support rom the Nanyang SUG Grant (project number M52000) and Mcrosot Operatons PE D-NU Jont R&D (project number M420065). 6. REFERENCES []. otellng Analyss o a Comple o Statstcal Varables nto Prncpal Components Journal o Educatonal Psychology vol. 24 pp [2] R. A. Fsher he Use o Multple Measurements n aonomc Problems Annals o Eugencs vol. 7 pp [3] D. ao et al. General ensor Dscrmnant Analyss and Gabor Features or Gat Recognton IEEE rans. Pattern Analyss and Machne Intellgence vol. 29 no. 0 pp [4]. e and P. Nyog ocalty Preservng Projectons NIPS vol. 6 Vancouver Canada [5]. Zhang et al. Dscrmnatve ocalty Algnment Proc. European Con. Computer Vson vol. pp [6]. Zhang et al. Patch algnment or dmensonalty reducton IEEE rans. Knowledge and Data Engneerng [7] D.. Donoho and C. Grmes essan Egenmaps: new locally lnear embeddng technques or hgh-dmensonal data Proc. Natonal Academy o Arts and Scences vol. 00 pp [8]. Zhang et al. A unyng ramework or spectral analyss based dmensonalty reducton Int l Jont Con. Neural Networks pp [9]. Sm et al. he CMU Pose Illumnaton and Epresson (PIE) Database o uman Faces echncal Report CMU-RI- R-0-02 Carnege Mellon Unversty 200 [0] D. P. Bertsekas Constraned Optmzaton and agrange Multpler (Optmzaton and Neural Computaton Seres) Athena Scentc 996. [] D. ao et al. Geometrc Mean or Subspace Selecton IEEE rans. Pattern Analyss and Machne Intellgence vol. 3 no. 2 pp [2] D. u et al. Margnal Fsher Analyss and Its Varants or uman Gat Recognton and Content Based Image Retreval IEEE rans. Image Processng vol. 6 no. pp [3] W. u et al. ransductve Component Analyss Proc. IEEE Int l Con. Data Mnng pp
Feature Reduction and Selection
Feature Reducton and Selecton Dr. Shuang LIANG School of Software Engneerng TongJ Unversty Fall, 2012 Today s Topcs Introducton Problems of Dmensonalty Feature Reducton Statstc methods Prncpal Components
More informationFace Recognition University at Buffalo CSE666 Lecture Slides Resources:
Face Recognton Unversty at Buffalo CSE666 Lecture Sldes Resources: http://www.face-rec.org/algorthms/ Overvew of face recognton algorthms Correlaton - Pxel based correspondence between two face mages Structural
More informationLECTURE : MANIFOLD LEARNING
LECTURE : MANIFOLD LEARNING Rta Osadchy Some sldes are due to L.Saul, V. C. Raykar, N. Verma Topcs PCA MDS IsoMap LLE EgenMaps Done! Dmensonalty Reducton Data representaton Inputs are real-valued vectors
More informationClassifier Selection Based on Data Complexity Measures *
Classfer Selecton Based on Data Complexty Measures * Edth Hernández-Reyes, J.A. Carrasco-Ochoa, and J.Fco. Martínez-Trndad Natonal Insttute for Astrophyscs, Optcs and Electroncs, Lus Enrque Erro No.1 Sta.
More informationSupport Vector Machines
/9/207 MIST.6060 Busness Intellgence and Data Mnng What are Support Vector Machnes? Support Vector Machnes Support Vector Machnes (SVMs) are supervsed learnng technques that analyze data and recognze patterns.
More informationTensor Locality Preserving Projections Based Urban Building Areas Extraction from High-Resolution SAR Images
Journal o Advances n Inormaton Technology Vol. 7, No. 4, November 016 Tensor Localty Preservng Proectons Based Urban Buldng Areas Extracton rom Hgh-Resoluton SAR Images Bo Cheng, Sha Cu, and Tng L Insttute
More informationSemi-Supervised Discriminant Analysis Based On Data Structure
IOSR Journal of Computer Engneerng (IOSR-JCE) e-issn: 2278-0661,p-ISSN: 2278-8727, Volume 17, Issue 3, Ver. VII (May Jun. 2015), PP 39-46 www.osrournals.org Sem-Supervsed Dscrmnant Analyss Based On Data
More informationSupport Vector Machines
Support Vector Machnes Decson surface s a hyperplane (lne n 2D) n feature space (smlar to the Perceptron) Arguably, the most mportant recent dscovery n machne learnng In a nutshell: map the data to a predetermned
More informationFace Recognition Based on SVM and 2DPCA
Vol. 4, o. 3, September, 2011 Face Recognton Based on SVM and 2DPCA Tha Hoang Le, Len Bu Faculty of Informaton Technology, HCMC Unversty of Scence Faculty of Informaton Scences and Engneerng, Unversty
More informationMULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION
MULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION Paulo Quntlano 1 & Antono Santa-Rosa 1 Federal Polce Department, Brasla, Brazl. E-mals: quntlano.pqs@dpf.gov.br and
More informationLaplacian Eigenmap for Image Retrieval
Laplacan Egenmap for Image Retreval Xaofe He Partha Nyog Department of Computer Scence The Unversty of Chcago, 1100 E 58 th Street, Chcago, IL 60637 ABSTRACT Dmensonalty reducton has been receved much
More informationDiscriminative Dictionary Learning with Pairwise Constraints
Dscrmnatve Dctonary Learnng wth Parwse Constrants Humn Guo Zhuoln Jang LARRY S. DAVIS UNIVERSITY OF MARYLAND Nov. 6 th, Outlne Introducton/motvaton Dctonary Learnng Dscrmnatve Dctonary Learnng wth Parwse
More informationParallelism for Nested Loops with Non-uniform and Flow Dependences
Parallelsm for Nested Loops wth Non-unform and Flow Dependences Sam-Jn Jeong Dept. of Informaton & Communcaton Engneerng, Cheonan Unversty, 5, Anseo-dong, Cheonan, Chungnam, 330-80, Korea. seong@cheonan.ac.kr
More informationLearning the Kernel Parameters in Kernel Minimum Distance Classifier
Learnng the Kernel Parameters n Kernel Mnmum Dstance Classfer Daoqang Zhang 1,, Songcan Chen and Zh-Hua Zhou 1* 1 Natonal Laboratory for Novel Software Technology Nanjng Unversty, Nanjng 193, Chna Department
More informationAnnouncements. Supervised Learning
Announcements See Chapter 5 of Duda, Hart, and Stork. Tutoral by Burge lnked to on web page. Supervsed Learnng Classfcaton wth labeled eamples. Images vectors n hgh-d space. Supervsed Learnng Labeled eamples
More informationHigh Dimensional Data Clustering
Hgh Dmensonal Data Clusterng Charles Bouveyron 1,2, Stéphane Grard 1, and Cordela Schmd 2 1 LMC-IMAG, BP 53, Unversté Grenoble 1, 38041 Grenoble Cede 9, France charles.bouveyron@mag.fr, stephane.grard@mag.fr
More informationRange images. Range image registration. Examples of sampling patterns. Range images and range surfaces
Range mages For many structured lght scanners, the range data forms a hghly regular pattern known as a range mage. he samplng pattern s determned by the specfc scanner. Range mage regstraton 1 Examples
More informationCluster Analysis of Electrical Behavior
Journal of Computer and Communcatons, 205, 3, 88-93 Publshed Onlne May 205 n ScRes. http://www.scrp.org/ournal/cc http://dx.do.org/0.4236/cc.205.350 Cluster Analyss of Electrcal Behavor Ln Lu Ln Lu, School
More informationThe Discriminate Analysis and Dimension Reduction Methods of High Dimension
Open Journal of Socal Scences, 015, 3, 7-13 Publshed Onlne March 015 n ScRes. http://www.scrp.org/journal/jss http://dx.do.org/10.436/jss.015.3300 The Dscrmnate Analyss and Dmenson Reducton Methods of
More informationContent Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers
IOSR Journal of Electroncs and Communcaton Engneerng (IOSR-JECE) e-issn: 78-834,p- ISSN: 78-8735.Volume 9, Issue, Ver. IV (Mar - Apr. 04), PP 0-07 Content Based Image Retreval Usng -D Dscrete Wavelet wth
More informationRecognizing Faces. Outline
Recognzng Faces Drk Colbry Outlne Introducton and Motvaton Defnng a feature vector Prncpal Component Analyss Lnear Dscrmnate Analyss !"" #$""% http://www.nfotech.oulu.f/annual/2004 + &'()*) '+)* 2 ! &
More information12/2/2009. Announcements. Parametric / Non-parametric. Case-Based Reasoning. Nearest-Neighbor on Images. Nearest-Neighbor Classification
Introducton to Artfcal Intellgence V22.0472-001 Fall 2009 Lecture 24: Nearest-Neghbors & Support Vector Machnes Rob Fergus Dept of Computer Scence, Courant Insttute, NYU Sldes from Danel Yeung, John DeNero
More informationMachine Learning 9. week
Machne Learnng 9. week Mappng Concept Radal Bass Functons (RBF) RBF Networks 1 Mappng It s probably the best scenaro for the classfcaton of two dataset s to separate them lnearly. As you see n the below
More informationCS246: Mining Massive Datasets Jure Leskovec, Stanford University
CS46: Mnng Massve Datasets Jure Leskovec, Stanford Unversty http://cs46.stanford.edu /19/013 Jure Leskovec, Stanford CS46: Mnng Massve Datasets, http://cs46.stanford.edu Perceptron: y = sgn( x Ho to fnd
More informationClassification / Regression Support Vector Machines
Classfcaton / Regresson Support Vector Machnes Jeff Howbert Introducton to Machne Learnng Wnter 04 Topcs SVM classfers for lnearly separable classes SVM classfers for non-lnearly separable classes SVM
More informationData-dependent Hashing Based on p-stable Distribution
Data-depent Hashng Based on p-stable Dstrbuton Author Ba, Xao, Yang, Hachuan, Zhou, Jun, Ren, Peng, Cheng, Jan Publshed 24 Journal Ttle IEEE Transactons on Image Processng DOI https://do.org/.9/tip.24.2352458
More informationSubspace clustering. Clustering. Fundamental to all clustering techniques is the choice of distance measure between data points;
Subspace clusterng Clusterng Fundamental to all clusterng technques s the choce of dstance measure between data ponts; D q ( ) ( ) 2 x x = x x, j k = 1 k jk Squared Eucldean dstance Assumpton: All features
More informationSmoothing Spline ANOVA for variable screening
Smoothng Splne ANOVA for varable screenng a useful tool for metamodels tranng and mult-objectve optmzaton L. Rcco, E. Rgon, A. Turco Outlne RSM Introducton Possble couplng Test case MOO MOO wth Game Theory
More informationSum of Linear and Fractional Multiobjective Programming Problem under Fuzzy Rules Constraints
Australan Journal of Basc and Appled Scences, 2(4): 1204-1208, 2008 ISSN 1991-8178 Sum of Lnear and Fractonal Multobjectve Programmng Problem under Fuzzy Rules Constrants 1 2 Sanjay Jan and Kalash Lachhwan
More informationUsing Neural Networks and Support Vector Machines in Data Mining
Usng eural etworks and Support Vector Machnes n Data Mnng RICHARD A. WASIOWSKI Computer Scence Department Calforna State Unversty Domnguez Hlls Carson, CA 90747 USA Abstract: - Multvarate data analyss
More informationThe Research of Support Vector Machine in Agricultural Data Classification
The Research of Support Vector Machne n Agrcultural Data Classfcaton Le Sh, Qguo Duan, Xnmng Ma, Me Weng College of Informaton and Management Scence, HeNan Agrcultural Unversty, Zhengzhou 45000 Chna Zhengzhou
More informationTwo-Dimensional Supervised Discriminant Projection Method For Feature Extraction
Appl. Math. Inf. c. 6 No. pp. 8-85 (0) Appled Mathematcs & Informaton cences An Internatonal Journal @ 0 NP Natural cences Publshng Cor. wo-dmensonal upervsed Dscrmnant Proecton Method For Feature Extracton
More informationLearning a Locality Preserving Subspace for Visual Recognition
Learnng a Localty Preservng Subspace for Vsual Recognton Xaofe He *, Shucheng Yan #, Yuxao Hu, and Hong-Jang Zhang Mcrosoft Research Asa, Bejng 100080, Chna * Department of Computer Scence, Unversty of
More informationFace Recognition Method Based on Within-class Clustering SVM
Face Recognton Method Based on Wthn-class Clusterng SVM Yan Wu, Xao Yao and Yng Xa Department of Computer Scence and Engneerng Tong Unversty Shangha, Chna Abstract - A face recognton method based on Wthn-class
More informationClassification of Face Images Based on Gender using Dimensionality Reduction Techniques and SVM
Classfcaton of Face Images Based on Gender usng Dmensonalty Reducton Technques and SVM Fahm Mannan 260 266 294 School of Computer Scence McGll Unversty Abstract Ths report presents gender classfcaton based
More informationFacial Expression Recognition Based on Local Binary Patterns and Local Fisher Discriminant Analysis
WSEAS RANSACIONS on SIGNAL PROCESSING Shqng Zhang, Xaomng Zhao, Bcheng Le Facal Expresson Recognton Based on Local Bnary Patterns and Local Fsher Dscrmnant Analyss SHIQING ZHANG, XIAOMING ZHAO, BICHENG
More informationOptimal Design of Nonlinear Fuzzy Model by Means of Independent Fuzzy Scatter Partition
Optmal Desgn of onlnear Fuzzy Model by Means of Independent Fuzzy Scatter Partton Keon-Jun Park, Hyung-Kl Kang and Yong-Kab Km *, Department of Informaton and Communcaton Engneerng, Wonkwang Unversty,
More informationGender Classification using Interlaced Derivative Patterns
Gender Classfcaton usng Interlaced Dervatve Patterns Author Shobernejad, Ameneh, Gao, Yongsheng Publshed 2 Conference Ttle Proceedngs of the 2th Internatonal Conference on Pattern Recognton (ICPR 2) DOI
More informationImage Alignment CSC 767
Image Algnment CSC 767 Image algnment Image from http://graphcs.cs.cmu.edu/courses/15-463/2010_fall/ Image algnment: Applcatons Panorama sttchng Image algnment: Applcatons Recognton of object nstances
More informationLearning an Image Manifold for Retrieval
Learnng an Image Manfold for Retreval Xaofe He*, We-Yng Ma, and Hong-Jang Zhang Mcrosoft Research Asa Bejng, Chna, 100080 {wyma,hjzhang}@mcrosoft.com *Department of Computer Scence, The Unversty of Chcago
More informationA PATTERN RECOGNITION APPROACH TO IMAGE SEGMENTATION
1 THE PUBLISHING HOUSE PROCEEDINGS OF THE ROMANIAN ACADEMY, Seres A, OF THE ROMANIAN ACADEMY Volume 4, Number 2/2003, pp.000-000 A PATTERN RECOGNITION APPROACH TO IMAGE SEGMENTATION Tudor BARBU Insttute
More informationProblem Definitions and Evaluation Criteria for Computational Expensive Optimization
Problem efntons and Evaluaton Crtera for Computatonal Expensve Optmzaton B. Lu 1, Q. Chen and Q. Zhang 3, J. J. Lang 4, P. N. Suganthan, B. Y. Qu 6 1 epartment of Computng, Glyndwr Unversty, UK Faclty
More informationOutline. Discriminative classifiers for image recognition. Where in the World? A nearest neighbor recognition example 4/14/2011. CS 376 Lecture 22 1
4/14/011 Outlne Dscrmnatve classfers for mage recognton Wednesday, Aprl 13 Krsten Grauman UT-Austn Last tme: wndow-based generc obect detecton basc ppelne face detecton wth boostng as case study Today:
More informationA Binarization Algorithm specialized on Document Images and Photos
A Bnarzaton Algorthm specalzed on Document mages and Photos Ergna Kavalleratou Dept. of nformaton and Communcaton Systems Engneerng Unversty of the Aegean kavalleratou@aegean.gr Abstract n ths paper, a
More informationMULTI-VIEW ANCHOR GRAPH HASHING
MULTI-VIEW ANCHOR GRAPH HASHING Saehoon Km 1 and Seungjn Cho 1,2 1 Department of Computer Scence and Engneerng, POSTECH, Korea 2 Dvson of IT Convergence Engneerng, POSTECH, Korea {kshkawa, seungjn}@postech.ac.kr
More informationInfrared face recognition using texture descriptors
Infrared face recognton usng texture descrptors Moulay A. Akhlouf*, Abdelhakm Bendada Computer Vson and Systems Laboratory, Laval Unversty, Quebec, QC, Canada G1V0A6 ABSTRACT Face recognton s an area of
More informationOutline. Self-Organizing Maps (SOM) US Hebbian Learning, Cntd. The learning rule is Hebbian like:
Self-Organzng Maps (SOM) Turgay İBRİKÇİ, PhD. Outlne Introducton Structures of SOM SOM Archtecture Neghborhoods SOM Algorthm Examples Summary 1 2 Unsupervsed Hebban Learnng US Hebban Learnng, Cntd 3 A
More informationPositive Semi-definite Programming Localization in Wireless Sensor Networks
Postve Sem-defnte Programmng Localzaton n Wreless Sensor etworks Shengdong Xe 1,, Jn Wang, Aqun Hu 1, Yunl Gu, Jang Xu, 1 School of Informaton Scence and Engneerng, Southeast Unversty, 10096, anjng Computer
More informationAn Optimal Algorithm for Prufer Codes *
J. Software Engneerng & Applcatons, 2009, 2: 111-115 do:10.4236/jsea.2009.22016 Publshed Onlne July 2009 (www.scrp.org/journal/jsea) An Optmal Algorthm for Prufer Codes * Xaodong Wang 1, 2, Le Wang 3,
More informationCompetitive Sparse Representation Classification for Face Recognition
Vol. 6, No. 8, 05 Compettve Sparse Representaton Classfcaton for Face Recognton Yng Lu Chongqng Key Laboratory of Computatonal Intellgence Chongqng Unversty of Posts and elecommuncatons Chongqng, Chna
More informationIEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 1. SSDH: Semi-supervised Deep Hashing for Large Scale Image Retrieval
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY SSDH: Sem-supervsed Deep Hashng for Large Scale Image Retreval Jan Zhang, and Yuxn Peng arxv:607.08477v2 [cs.cv] 8 Jun 207 Abstract Hashng
More informationNUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS
ARPN Journal of Engneerng and Appled Scences 006-017 Asan Research Publshng Network (ARPN). All rghts reserved. NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS Igor Grgoryev, Svetlana
More informationSupport Vector Machines. CS534 - Machine Learning
Support Vector Machnes CS534 - Machne Learnng Perceptron Revsted: Lnear Separators Bnar classfcaton can be veed as the task of separatng classes n feature space: b > 0 b 0 b < 0 f() sgn( b) Lnear Separators
More informationLecture 4: Principal components
/3/6 Lecture 4: Prncpal components 3..6 Multvarate lnear regresson MLR s optmal for the estmaton data...but poor for handlng collnear data Covarance matrx s not nvertble (large condton number) Robustness
More informationIntegrated Expression-Invariant Face Recognition with Constrained Optical Flow
Integrated Expresson-Invarant Face Recognton wth Constraned Optcal Flow Chao-Kue Hseh, Shang-Hong La 2, and Yung-Chang Chen Department of Electrcal Engneerng, Natonal Tsng Hua Unversty, Tawan 2 Department
More informationMachine Learning: Algorithms and Applications
14/05/1 Machne Learnng: Algorthms and Applcatons Florano Zn Free Unversty of Bozen-Bolzano Faculty of Computer Scence Academc Year 011-01 Lecture 10: 14 May 01 Unsupervsed Learnng cont Sldes courtesy of
More informationTone-Aware Sparse Representation for Face Recognition
Tone-Aware Sparse Representaton for Face Recognton Lngfeng Wang, Huayu Wu and Chunhong Pan Abstract It s stll a very challengng task to recognze a face n a real world scenaro, snce the face may be corrupted
More informationCollaboratively Regularized Nearest Points for Set Based Recognition
Academc Center for Computng and Meda Studes, Kyoto Unversty Collaboratvely Regularzed Nearest Ponts for Set Based Recognton Yang Wu, Mchhko Mnoh, Masayuk Mukunok Kyoto Unversty 9/1/013 BMVC 013 @ Brstol,
More informationDetection of an Object by using Principal Component Analysis
Detecton of an Object by usng Prncpal Component Analyss 1. G. Nagaven, 2. Dr. T. Sreenvasulu Reddy 1. M.Tech, Department of EEE, SVUCE, Trupath, Inda. 2. Assoc. Professor, Department of ECE, SVUCE, Trupath,
More informationA Novel Term_Class Relevance Measure for Text Categorization
A Novel Term_Class Relevance Measure for Text Categorzaton D S Guru, Mahamad Suhl Department of Studes n Computer Scence, Unversty of Mysore, Mysore, Inda Abstract: In ths paper, we ntroduce a new measure
More informationOptimal Multiscale Organization of Multimedia Content for Fast Browsing and Cost-Effective Transmission
Proceedngs o the 6th WSEAS Internatonal Conerence on Sgnal Processng, Robotcs and Automaton, Coru Island, reece, February 6-9, 2007 257 Optmal Multscale Organzaton o Multmeda Content or Fast Browsng and
More informationMachine Learning. Support Vector Machines. (contains material adapted from talks by Constantin F. Aliferis & Ioannis Tsamardinos, and Martin Law)
Machne Learnng Support Vector Machnes (contans materal adapted from talks by Constantn F. Alfers & Ioanns Tsamardnos, and Martn Law) Bryan Pardo, Machne Learnng: EECS 349 Fall 2014 Support Vector Machnes
More informationUnsupervised Learning and Clustering
Unsupervsed Learnng and Clusterng Supervsed vs. Unsupervsed Learnng Up to now we consdered supervsed learnng scenaro, where we are gven 1. samples 1,, n 2. class labels for all samples 1,, n Ths s also
More informationSELECTION OF THE NUMBER OF NEIGHBOURS OF EACH DATA POINT FOR THE LOCALLY LINEAR EMBEDDING ALGORITHM
ISSN 392 24X INFORMATION TECHNOLOGY AND CONTROL, 2007, Vol.36, No.4 SELECTION OF THE NUMBER OF NEIGHBOURS OF EACH DATA POINT FOR THE LOCALLY LINEAR EMBEDDING ALGORITHM Rasa Karbauskatė,2, Olga Kurasova,2,
More informationBAYESIAN MULTI-SOURCE DOMAIN ADAPTATION
BAYESIAN MULTI-SOURCE DOMAIN ADAPTATION SHI-LIANG SUN, HONG-LEI SHI Department of Computer Scence and Technology, East Chna Normal Unversty 500 Dongchuan Road, Shangha 200241, P. R. Chna E-MAIL: slsun@cs.ecnu.edu.cn,
More informationAppearance-based Statistical Methods for Face Recognition
47th Internatonal Symposum ELMAR-2005, 08-10 June 2005, Zadar, Croata Appearance-based Statstcal Methods for Face Recognton Kresmr Delac 1, Mslav Grgc 2, Panos Latss 3 1 Croatan elecom, Savsa 32, Zagreb,
More informationRECOGNIZING GENDER THROUGH FACIAL IMAGE USING SUPPORT VECTOR MACHINE
Journal of Theoretcal and Appled Informaton Technology 30 th June 06. Vol.88. No.3 005-06 JATIT & LLS. All rghts reserved. ISSN: 99-8645 www.jatt.org E-ISSN: 87-395 RECOGNIZING GENDER THROUGH FACIAL IMAGE
More informationA Novel Fingerprint Matching Method Combining Geometric and Texture Features
A Novel ngerprnt Matchng Method Combnng Geometrc and Texture eatures Me Xe, Chengpu Yu and Jn Q Unversty of Electronc Scence and Technology of Chna. Chengdu,P.R.Chna xeme@ee.uestc.edu.cn Post Code:6154
More informationTerm Weighting Classification System Using the Chi-square Statistic for the Classification Subtask at NTCIR-6 Patent Retrieval Task
Proceedngs of NTCIR-6 Workshop Meetng, May 15-18, 2007, Tokyo, Japan Term Weghtng Classfcaton System Usng the Ch-square Statstc for the Classfcaton Subtask at NTCIR-6 Patent Retreval Task Kotaro Hashmoto
More informationA Bilinear Model for Sparse Coding
A Blnear Model for Sparse Codng Davd B. Grmes and Rajesh P. N. Rao Department of Computer Scence and Engneerng Unversty of Washngton Seattle, WA 98195-2350, U.S.A. grmes,rao @cs.washngton.edu Abstract
More informationFEATURE EXTRACTION. Dr. K.Vijayarekha. Associate Dean School of Electrical and Electronics Engineering SASTRA University, Thanjavur
FEATURE EXTRACTION Dr. K.Vjayarekha Assocate Dean School of Electrcal and Electroncs Engneerng SASTRA Unversty, Thanjavur613 41 Jont Intatve of IITs and IISc Funded by MHRD Page 1 of 8 Table of Contents
More informationOutline. Type of Machine Learning. Examples of Application. Unsupervised Learning
Outlne Artfcal Intellgence and ts applcatons Lecture 8 Unsupervsed Learnng Professor Danel Yeung danyeung@eee.org Dr. Patrck Chan patrckchan@eee.org South Chna Unversty of Technology, Chna Introducton
More informationThe Research of Ellipse Parameter Fitting Algorithm of Ultrasonic Imaging Logging in the Casing Hole
Appled Mathematcs, 04, 5, 37-3 Publshed Onlne May 04 n ScRes. http://www.scrp.org/journal/am http://dx.do.org/0.436/am.04.584 The Research of Ellpse Parameter Fttng Algorthm of Ultrasonc Imagng Loggng
More informationGraph-based Clustering
Graphbased Clusterng Transform the data nto a graph representaton ertces are the data ponts to be clustered Edges are eghted based on smlarty beteen data ponts Graph parttonng Þ Each connected component
More informationCHAPTER 3 SEQUENTIAL MINIMAL OPTIMIZATION TRAINED SUPPORT VECTOR CLASSIFIER FOR CANCER PREDICTION
48 CHAPTER 3 SEQUENTIAL MINIMAL OPTIMIZATION TRAINED SUPPORT VECTOR CLASSIFIER FOR CANCER PREDICTION 3.1 INTRODUCTION The raw mcroarray data s bascally an mage wth dfferent colors ndcatng hybrdzaton (Xue
More informationWavelets and Support Vector Machines for Texture Classification
Wavelets and Support Vector Machnes for Texture Classfcaton Kashf Mahmood Rapoot Faculty of Computer Scence & Engneerng, Ghulam Ishaq Khan Insttute, Top, PAKISTAN. kmr@gk.edu.pk Nasr Mahmood Rapoot Department
More informationCLASSIFICATION OF ULTRASONIC SIGNALS
The 8 th Internatonal Conference of the Slovenan Socety for Non-Destructve Testng»Applcaton of Contemporary Non-Destructve Testng n Engneerng«September -3, 5, Portorož, Slovena, pp. 7-33 CLASSIFICATION
More informationNonlocal Mumford-Shah Model for Image Segmentation
for Image Segmentaton 1 College of Informaton Engneerng, Qngdao Unversty, Qngdao, 266000,Chna E-mal:ccluxaoq@163.com ebo e 23 College of Informaton Engneerng, Qngdao Unversty, Qngdao, 266000,Chna E-mal:
More informationSHAPE RECOGNITION METHOD BASED ON THE k-nearest NEIGHBOR RULE
SHAPE RECOGNITION METHOD BASED ON THE k-nearest NEIGHBOR RULE Dorna Purcaru Faculty of Automaton, Computers and Electroncs Unersty of Craoa 13 Al. I. Cuza Street, Craoa RO-1100 ROMANIA E-mal: dpurcaru@electroncs.uc.ro
More informationEnhanced AMBTC for Image Compression using Block Classification and Interpolation
Internatonal Journal of Computer Applcatons (0975 8887) Volume 5 No.0, August 0 Enhanced AMBTC for Image Compresson usng Block Classfcaton and Interpolaton S. Vmala Dept. of Comp. Scence Mother Teresa
More informationTransductive Regression Piloted by Inter-Manifold Relations
Huan Wang IE, The Chnese Unversty of Hong Kong, Hong Kong Shucheng Yan Thomas Huang ECE, Unversty of Illnos at Urbana Champagn, USA Janzhuang Lu Xaoou Tang IE, The Chnese Unversty of Hong Kong, Hong Kong
More informationMulti-stable Perception. Necker Cube
Mult-stable Percepton Necker Cube Spnnng dancer lluson, Nobuuk Kaahara Fttng and Algnment Computer Vson Szelsk 6.1 James Has Acknowledgment: Man sldes from Derek Hoem, Lana Lazebnk, and Grauman&Lebe 2008
More informationCorner-Based Image Alignment using Pyramid Structure with Gradient Vector Similarity
Journal of Sgnal and Informaton Processng, 013, 4, 114-119 do:10.436/jsp.013.43b00 Publshed Onlne August 013 (http://www.scrp.org/journal/jsp) Corner-Based Image Algnment usng Pyramd Structure wth Gradent
More informationPCA Based Gait Segmentation
Honggu L, Cupng Sh & Xngguo L PCA Based Gat Segmentaton PCA Based Gat Segmentaton Honggu L, Cupng Sh, and Xngguo L 2 Electronc Department, Physcs College, Yangzhou Unversty, 225002 Yangzhou, Chna 2 Department
More informationComputer Science Technical Report
Computer Scence echncal Report NLYSIS OF PCSED ND FISHER DISCRIMINNSED IMGE RECOGNIION LGORIHMS Wendy S. Yambor July echncal Report CS3 Computer Scence Department Colorado State Unversty Fort Collns, CO
More informationControl strategies for network efficiency and resilience with route choice
Control strateges for networ effcency and reslence wth route choce Andy Chow Ru Sha Centre for Transport Studes Unversty College London, UK Centralsed strateges UK 1 Centralsed strateges Some effectve
More informationIntra-Parametric Analysis of a Fuzzy MOLP
Intra-Parametrc Analyss of a Fuzzy MOLP a MIAO-LING WANG a Department of Industral Engneerng and Management a Mnghsn Insttute of Technology and Hsnchu Tawan, ROC b HSIAO-FAN WANG b Insttute of Industral
More informationSLAM Summer School 2006 Practical 2: SLAM using Monocular Vision
SLAM Summer School 2006 Practcal 2: SLAM usng Monocular Vson Javer Cvera, Unversty of Zaragoza Andrew J. Davson, Imperal College London J.M.M Montel, Unversty of Zaragoza. josemar@unzar.es, jcvera@unzar.es,
More informationEcient Computation of the Most Probable Motion from Fuzzy. Moshe Ben-Ezra Shmuel Peleg Michael Werman. The Hebrew University of Jerusalem
Ecent Computaton of the Most Probable Moton from Fuzzy Correspondences Moshe Ben-Ezra Shmuel Peleg Mchael Werman Insttute of Computer Scence The Hebrew Unversty of Jerusalem 91904 Jerusalem, Israel Emal:
More informationKernel Collaborative Representation Classification Based on Adaptive Dictionary Learning
Internatonal Journal of Intellgent Informaton Systems 2018; 7(2): 15-22 http://www.scencepublshnggroup.com/j/js do: 10.11648/j.js.20180702.11 ISSN: 2328-7675 (Prnt); ISSN: 2328-7683 (Onlne) Kernel Collaboratve
More informationAn Improved Image Segmentation Algorithm Based on the Otsu Method
3th ACIS Internatonal Conference on Software Engneerng, Artfcal Intellgence, Networkng arallel/dstrbuted Computng An Improved Image Segmentaton Algorthm Based on the Otsu Method Mengxng Huang, enjao Yu,
More informationAbstract metric to nd the optimal pose and to measure the distance between the measurements
3D Dstance Metrc for Pose Estmaton and Object Recognton from 2D Projectons Yacov Hel-Or The Wezmann Insttute of Scence Dept. of Appled Mathematcs and Computer Scence Rehovot 761, ISRAEL emal:toky@wsdom.wezmann.ac.l
More informationNon-Negative Matrix Factorization and Support Vector Data Description Based One Class Classification
IJCSI Internatonal Journal of Computer Scence Issues, Vol. 9, Issue 5, No, September 01 ISSN (Onlne): 1694-0814 www.ijcsi.org 36 Non-Negatve Matrx Factorzaton and Support Vector Data Descrpton Based One
More informationUnsupervised Learning
Pattern Recognton Lecture 8 Outlne Introducton Unsupervsed Learnng Parametrc VS Non-Parametrc Approach Mxture of Denstes Maxmum-Lkelhood Estmates Clusterng Prof. Danel Yeung School of Computer Scence and
More informationFeature Extraction Based on Maximum Nearest Subspace Margin Criterion
Neural Process Lett DOI 10.7/s11063-012-9252-y Feature Extracton Based on Maxmum Nearest Subspace Margn Crteron Y Chen Zhenzhen L Zhong Jn Sprnger Scence+Busness Meda New York 2012 Abstract Based on the
More informationCS434a/541a: Pattern Recognition Prof. Olga Veksler. Lecture 15
CS434a/541a: Pattern Recognton Prof. Olga Veksler Lecture 15 Today New Topc: Unsupervsed Learnng Supervsed vs. unsupervsed learnng Unsupervsed learnng Net Tme: parametrc unsupervsed learnng Today: nonparametrc
More informationObject Recognition Based on Photometric Alignment Using Random Sample Consensus
Vol. 44 No. SIG 9(CVIM 7) July 2003 3 attached shadow photometrc algnment RANSAC RANdom SAmple Consensus Yale Face Database B RANSAC Object Recognton Based on Photometrc Algnment Usng Random Sample Consensus
More informationFace Recognition using 3D Directional Corner Points
2014 22nd Internatonal Conference on Pattern Recognton Face Recognton usng 3D Drectonal Corner Ponts Xun Yu, Yongsheng Gao School of Engneerng Grffth Unversty Nathan, QLD, Australa xun.yu@grffthun.edu.au,
More informationHistogram-Enhanced Principal Component Analysis for Face Recognition
Hstogram-Enhanced Prncpal Component Analyss for Face ecognton Ana-ara Sevcenco and Wu-Sheng Lu Dept. of Electrcal and Computer Engneerng Unversty of Vctora sevcenco@engr.uvc.ca, wslu@ece.uvc.ca Abstract
More informationCalibrating a single camera. Odilon Redon, Cyclops, 1914
Calbratng a sngle camera Odlon Redon, Cclops, 94 Our goal: Recover o 3D structure Recover o structure rom one mage s nherentl ambguous??? Sngle-vew ambgut Sngle-vew ambgut Rashad Alakbarov shadow sculptures
More information