Tone-Aware Sparse Representation for Face Recognition

Size: px
Start display at page:

Download "Tone-Aware Sparse Representation for Face Recognition"

Transcription

1 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 by many unknown factors. Among them, llumnaton varaton s an mportant one, whch wll be manly dscussed n ths paper. Frst, the llumnaton varatons caused by shadow or overexposure are regarded as a multplcatve scalng mage over the orgnal face mage. The purpose of ntroducng scalng vector (or scalng mage) s to enhance the pxels n shadow regons, whle depress the pxels n overexposure regons. Then, based on the scalng vector, we propose a novel tone-aware sparse representaton (TASR) model. Fnally, a EMlke algorthm s proposed to solve the proposed TASR model. Extensve experments on the benchmark face databases show that our method s more robust aganst llumnaton varatons. I. INTRODUCTION Face recognton [14], [26] plays an mportant role n many computer vson applcatons, such as access control and securty survellance. Nevertheless, t s stll a challengng problem, snce face mages may be corrupted by many unknown factors, e.g., llumnaton, occluson, expresson, and agng. In the past decades, a number of researchers have been attracted to tackle these dffcultes. A. Prevous Work The face recognton approaches can be manly grouped nto two categores,.e., the subspace (manfold learnng) based and sparse representaton based methods. The tradtonal subspace (manfold learnng) based face recognton methods [19], [2], [22], [8], [3] are based on the assumpton that hgh dmensonal face mages should be le on lower dmensonal subspace or sub-manfold. So far, many subspace learnng (manfold learnng) methods have been successfully appled n face recognton. For example, the classcal Egen-face [19] and Fsherface [2] methods consder all face mage globally, and utlze prncpal component analyss or fsher lnear dscrmnant analyss to learn subspaces. To ntroduce local structures of all face mages, manfold learnng based methods, such as localty-preservng projecton [8] and local dscrmnant embeddng [3], have been proposed. Recently, the spare representaton (or sparse codng) based face recognton methods [20], [25], [23], [24], [7] have been pad much attentons. To the best of our knowledge, the earlest sparse representaton based face recognton has Ths work was supported by Natonal Natural Scence Foundaton of Chna (NSFC No , NSFC No , NSFC No and NSFC No ). L. Wang, H. Wu and C. Pan are wth the Department of NLPR, Insttute of Automaton, Chnese Academy of Scences, Bejng, Chna {lfwang,hywu,chpan}@nlpr.a.ac.cn Fg. 1. The overexposure and shadow examples obtaned from benchmark face databases. been proposed n [20]. In ths method, all tranng mages are frst collaborated together. Then, for each testng mage, t s sparsely represented over all tranng mages. Fnally, the classfcaton result s obtaned by fndng the class whch yelds least representaton error. Improved performances have been reported n [20], as compared wth the classcal subspace or manfold learnng based methods. To mprove robustness n case of occlusons, robust sparse representaton algorthms have been proposed n [24], [7] by ntroducng a weght for each pxel. The occluson pxels are assgned lower weghts, whle other pxels are assgned hgher weghts. The weghts and the codng vector are optmzed alternately. To mprove the raw ntensty feature used n the orgnal sparse representaton, [25], [10], [12], adopt LBP [16] feature, whle [23] uses Gabor [9] feature. In [21], both of features are used. B. Motvaton and the Proposed Method The orgnal sparse representaton based face recognton method can perform well when tranng and testng mages are all acqured n the well llumnaton condtons. Unfortunately, both mages may be corrupted by overexposure or shadow (see Fg. 1 for detal). In such cases, the tone of the mage s changed. To address ths problem, for each pxel on the testng mage, we ntroduce a scalng value. Based on ths, the test mage s transformed to a new tone-free verson, n whch the shadow regon can be enhanced whle the overexposure regon can be depressed. Then, by ntroducng ths tone-free mage nto the sparse representaton framework, we propose a novel tone-aware sparse representaton (TASR) model. The detals and advantages of our proposed TASR are lsted as follows: The proposed TASR model ncorporates tone varatons nto the sparse representaton. Hence, compared wth the classcal sparse representaton based face recognton, our TASR based face recognton s less senstve to llumnaton varatons on face mages. Extensve exper-

2 Fg. 2. An vsual descrpton of our tone-aware sparse representaton (TASR) model. The mages wth blue rectangles are known, whle the mages wth red rectangles are unknown. In our TASR model, the tone-free mage ȳ s sparsely represented by the nput tranng mages X. That s, our TASR need to estmate both the sparse codng vector α and the scalng vector s (scalng mage S). ments on the benchmark face databases also ndcate that our method provdes better adaptablty. An mage guded Laplacan regularzaton (IGLR) s presented to make the scalng mage locally smooth, whch s the matrx form of scalng vector. The man dea behnd IGLR s to transfer the local smoothness from gudance mage to a scalng mage by usng locally lnear regresson. The gudance mage s denoted as a medan face mage, n whch each pxel value s the medan flter result of the correspondng pxel values n the tranng mages. The expectaton maxmzaton lke (EM-lke) optmzaton algorthm s used to mnmze the TASR model. Based on ths algorthm, both the scalng vector and the sparse codng vector can be obtaned alternatvely. The remander of ths paper s organzed as follows. In Secton II, we brefly revew the orgnal sparse representaton based face recognton method. In Secton III, we descrbe our tone-aware sparse representaton (TASR) model n detal. Some expermental results are provded n Secton IV. The conclusve remarks are gven n Secton V. II. OVERVIEW OF ORIGINAL SPARSE REPRESENTATION Denote the tranng mages by X = [X 1, X 2,, X C ] R d n, where X s the subset of tranng mages of class, and C s the number of classes. Note that n s the number of tranng mages, and d = h w s feature length (or the number of pxels n each mage), where h and w are the heght and wdth of the mage, respectvely. For a gven testng mage y R d 1, the orgnal sparse representaton n [20] ams to obtan a codng vector α R n 1 by mnmzng followng equaton, gven by ˆα = mn α y Xα λ α 1, (1) where λ s a weghtng constant, whch balances the reconstructon error and the sparseness of codng vector. After obtanng the codng vector ˆα, the recognton s determned by fndng the least reconstructon error of all classes: IDENTITY(y) = arg mn y X ˆα, (2) where ˆα s the coeffcent vector assocated wth -th class, and ˆα = [ˆα 1 ; ˆα 2 ; ; ˆα C ]. III. TONE-AWARE SPARSE REPRESENTATION (TASR) To ntroduce tone, the testng mage y s transformed by multplyng a scalng vector s R d 1, gven by ȳ = y s, (3) where s the Hadamard product operator, and ȳ R d 1 s the transformed tone-free testng mage. The scalng vector s s used to depress or enhance face mage y, adaptvely. That s, for each pxel n the testng mage, f t s corrupted by shadow, the correspondng scalng value should be large (enhanced), whle f t s overexposured, the correspondng scalng value should be small (depressed). Our tone-aware sparse representaton model (TASR) makes the tone-free mage y s be sparsely represented by the tranng mages X, gven by ˆα, ŝ = mn α,s y s Xα λ α 1 + γreg(s), (4) where Reg(s) s the mage guded Laplacan regularzaton (IGLR) term to make the scalng vector be locally smooth, and parameter γ s a weghtng constant. Fg. 2 shows the flow chart of the proposed TASR n detal. It can be seen that the TASR need to nfer both the scalng vector (or the scalng mage) and the codng vector. In the followng, we frst descrbe the IGLR. Then, we present a expectaton maxmzaton lke (EM-lke) optmzaton algorthm to mnmze the proposed TASR model. A. Image Guded Laplacan Regularzaton (IGLR) To facltate the descrpton of Laplacan regularzaton, the scalng vector s s reshaped nto a h w matrx, called the scalng mage, whch s denoted as S R h w. In practce, the scalng mage S could be smooth locally, snce the shadow or overexposure regons on the testng face mage are pece-wse smooth. To ensure the smoothness of the scalng mage, we propose a new IGLR, n whch

3 the scalng mage S s locally lnear regressed by a known gudance mage G. Now, we consder a local regon (3 3 local wndow for example) Ω centered at -th pxel, namely Ω = { 0, 1, 2,..., 8 } (5) where 0 = s the ndex of -th pxel, and 1, 2,..., 8 are the ndexes of pxels arrangng from top-left to bottom-rght. In ths local regon, the scalng mage values are lnearly regressed by the gudance mage values, gven by S j = w G j + b, j = 0, 1, 2,..., 8, (6) where w and t are the regresson coeffcents n ths local regon Ω. To determne two regresson coeffcents w and b, we defne the followng cost functon: 8 E({S j } 8 ( j=0, w, t ) = Sj w G j b ) 2 + εw 2, (7) j=0 where ε s a small postve constant. The total cost functon s formulated by summng all local costs,.e., n 8 ( (Sj E(S, W, B) = w G j t ) ) 2 + εw 2, =1 j=0 (8) where W and B are the matrx formulatons of all regresson coeffcents. As proved n [13], mnmzng the cost functon E(S, W, B) wth respect to W and B equals to the Laplacan regularzaton s T Ls, that s, J(S) = mn W,B E(S, W, B) = st Ls. (9) Matrx L s a n n Laplacan matrx, whose (j, k)-th entry s ( δ jk 1 ( )) ε (G j µ )(G k µ ). σ2 { {j,k} Ω } Here, δ jk s the Kronecker delta functon, whle µ and σ are the mean and varance of the gudance mage G at the local regon Ω, respectvely. In ths paper, the proposed IGLR s defned as Reg(s) = s T Ls. (10) The IGLR s proposed to transfer the local smoothness from the gudance mage to the scalng mage. Hence, t s mportant to choose an approprate gudance mage G. In ths paper, the gudance mage G s set as the medan face mage, whch s obtaned by followng two steps. The -th entry of gudance vector g s frst calculated by g = med(x. ), (11) where X. the -th row of X, and med(.) s the medan flter operaton. Then, the gudance mage G s obtaned by reshapng the gudance vector g nto h w sze. Fg. 3 gves an example of a gudance mage obtaned from the AR face database. From ths fgure, t can be seen that the gudance mage s smooth, thus, makes the scalng mage also be smooth. Fg. 3. The gudance mage (or medan face mage) obtaned from the AR face database. The sub-fgure (b) gves a 3D vew result. B. Optmzaton of the Proposed TASR Model By substtutng the IGLR of Eqn. (10) nto Eqn. (4), we can rewrte the objectve functon TASR model as ˆα, ŝ = mn α,s y s Xα λ α 1 + γs T Ls + ξ s 1 2 2, (12) where ξ s weghtng constant and 1 s a n 1 vector wth all entres are 1. The term s s used to ensure scale values be close to 1. From Eqn. (12), t can be seen that t s very hard to obtan the codng vector α and the scalng vector s smultaneously. To tackle ths dffculty, a EM-lke method s utlzed by optmzng these two vectors alternately. Step 1: Fxng Scalng Vector s, Calculatng Codng Vector α. For a fxed scalng vector, the objectve functon of the TASR model s reformulated as ˆα = mn α ȳ Xα λ α 1, (13) where ȳ = y s s known. From Eqn. (13), we see that ths model s the same wth the orgnal sparse representaton model. Thereby, t can be solved by the tradtonal l 1 mnmzng methods, such as LASSO [18], FIST [1], LARS [5], and Homotopy [4]. In ths paper, Homotopy s utlzed to solve Eqn. (13), because t s less senstve to the choce of sparse regularzaton parameter λ. Please refer to [4] for more nformaton about Homotopy algorthm. Step 2: Fxng Codng Vector α, Calculatng Scalng Vector b. Keepng the codng vector, the scalng vector s obtaned by mnmzng the followng functon, gven by ŝ = mn s y s ŷ γs T Ls + ξ s 1 2 2, (14) where ŷ = Xα s known. To determne the scalng vector s, we frst reformulate Eqn. (14) as ŝ = mn s Ys ŷ γs T Ls + ξ s 1 2 2, (15) where Y s a n n dagonal matrx obtaned from vector y, namely, Y = dag(y). Therefore, the scalng vector can be obtaned by settng the dervatve of Eqn. (15) wth respect to s s equal to zero, gven by ŝ = (Y T Y + γl + ξi) 1 (Y T ŷ + ξ1), (16) where I s a dentty matrx. In ths method, the ntal scalng vector s assgned to the scalng vector obtaned n the prevous tme. Eqn. (16) s calculated by flterng method proposed n [6].

4 Fg. 4. Vsual comparson example of the AR database. (a), the testng mage. (b), the tone-free testng mage obtaned by our method. (c), seven tranng mages of the 35-th class. (d), seven tranng mages of the 47-th class. (e), the codng vector obtaned by the orgnal SR. (f), the codng vector obtaned by our TASR. The left-bottom mages of sub-fgures (e,f) are the testng mage and tone-free testng mage, respectvely. The testng mage belongs to the 35-th class. The orgnal SR based method msclassfes t as the 47-th class. Detals are gven n the text. C. Summarzaton of TASR Based Face Recognton As mentoned above, the codng vector α and the scalng vector s are obtaned alternately. Then, the recognton s determned by fndng the least reconstructon error of all classes: IDENTITY(y) = arg mn y ŝ X ˆα, (17) where ˆα s the coeffcent vector assocated wth -th class and ˆα = [ˆα 1 ; ˆα 2 ; ; ˆα C ]. Our TASR based face recognton algorthm s summarzed n Algorthm 1. IV. EXPERIMENTAL RESULTS Extensve experments on three publc face databases,.e., AR, Extended Yale B, and CMU-PIE, are performed to evaluate the effectveness of the proposed TASR based face recognton algorthm. We compare our algorthm wth the orgnal sparse representaton algorthm [20], snce t s most related to ours. For convenence, we denote sparse representaton method by SR for short. For our and the orgnal sparse representaton algorthms, the major parameter λ s set to a value of A. Results on AR wthout Occluson Database We frst test our TASR based face recognton algorthm on the AR wthout occluson database [15]. Same to [20], a subset of mages wth only llumnaton and expresson varatons are selected. For each object, seven mages that acqured n the frst sesson are selected for tranng, whle other seven mages that obtaned n the second sesson are selected for testng. All the mages, ncludng tranng and testng mages, are cropped and reszed to Algorthm 1: TASR Based Face Recognton Data: The tranng mage X, and testng mage y. Result: Identty result c. 1 Intalzng scalng vector s = 1, where 1 s a n 1 vector wth all entres are 1; 2 /* MAIN PROCEDURES OF TASR */ 3 for t 1 to maxiter do 4 Calculatng codng vector α t based on the Homotopy optmzaton of Eqn. (13); 5 Calculatng scalng vector s t based on Eqn. (16); 6 Calculatng reconstructon error e t = y s t Xα t ; 7 f e t e t 1 then 8 s t = s t 1 ; 9 α t = α t 1 ; break; end end /* MAIN PART OF FACE RECOGNITION */ c = arg mn y ŝ X ˆα ; The comparatve results are shown n Table I. From ths table, we see that the recognton rate of our method s hgher than that of the orgnal SR. In Fg. 4, we further present a vsual example to compare our TASR wth the orgnal SR. From ths fgure, we see that two persons are very smlar, snce both of them wear glasses. The testng mage belongs to the 35-th class. The orgnal SR msclassfes t as the 47-th class. It could be

5 TABLE I COMPARISONS ON THE AR WITHOUT OCCLUSION DATABASE Algorthm Recognton Rate SR TASR manly caused by the strong reflecton on the glass corner (see the red crcle on Fg. 4 (a)). Note that, the mages of the 47-th class also contan a sample wth strong reflecton on the glass corner (see the red crcle on Fg. 4 (d)). From Fg. 4 (b), we can see that the strong reflecton on the testng mage s weaken by our method, whch makes other regons, such as the mouth regon, become more salent. Fg. 4 (e) and (f) also verfy ths observaton. These two fgures show the codng vectors obtaned by the orgnal SR and our TASR, respectvely. As shown n sub-fgure (e), the codng value of the 324-th mage, whch belongs to 47-th class, s very hgh. However, as shown n Fg. 4 (f), the codng value of the 324 mage becomes lower, whle the codng value of the 240 mage, whch belongs to 35-th class, becomes larger, compared wth the orgnal codng vector of Fg. 4 (e). Although, the codng value of the 240-th mage s stll smaller than the codng value of the 324-th mage, t s suffcent to recognze the testng mage correctly wth the help of other mages belongng to the 35-th class. belongng to the 86-th class, whch further mprove the correct rate of our TASR method. B. Results on Extended YelaB Database The extended Yale B database [11] s consst of 16, 128 face mages of 38 subjects under 9 poses and 64 laboratorycontrolled llumnaton condtons. In ths experment, we select 2, 414 frontal mages. The resoluton of each mage s cropped and reszed nto Fg. 6 gves some examples. From ths fgure, we see that some face mages are greatly corrupted by the shadow. Fg. 6. Image samples from the extended Yale B database. Some face mages are corrupted by shadows. Table II presents the comparatve results on dfferent tranng samples. For each person, we randomly select 5, 10, 15, 20, 25 mages for tranng, and randomly select 300 mages from the rest mages for testng. For each fxed tranng and testng mages, we perform 10 tasks recognton for each algorthm. From table II, we see that the recognton rates of our method are all hgher than those of the SR. Fg. 7. Some face mages, whch are msclassfed by the orgnal SR method. The proposed TASR classfes these face mages correctly. Fg. 5. Another vsual comparson on the AR database. (a), the testng mage. (b), the tone-free testng mage obtaned by our method. (c), seven tranng mages of the 86-th class. (d), seven tranng mages of the 68- th class. The testng mage belongs to the 86-th class. The orgnal sparse representaton based method msclassfes t as the 68-th class. Fg. 5 llustrates another vsual comparson example. The orgnal SR based method msclassfes t as the 68-th class. The man reason s that both the llumnaton condton and the mouth expresson on the orgnal testng mage are both smlar to those on the ffth mage of the 68- th class (see the red rectangle). However, for the mages belongng to the 86-th class, they are dfferent from the testng mage ether n llumnaton condton or n the mouth expresson. Fortunately, by removng the overexposure, the tone-free testng mage could be more close to the the mages Fg. 7 presents some face mages, whch are msclassfed by the orgnal SR method, whle are classfed correctly by our TASR method. From ths fgure, we see that some of face mages are severely corrupted by shadows. For some face mages, t s even hard for human beng to recognze them. The success on these mages ndcates that our TASR model s more robust to shadow than the orgnal SR model. C. Results on CMU-PIE Database In the thrd experment, we compare our TASR wth the orgnal SR on the CMU-PIE database [17]. Ths face database has 41, 368 mages of 68 subjects under 13 dfferent poses, 43 dfferent llumnaton condtons, and 4 dfferent expressons. We select the frontal face mages n the experment. All mages are cropped and reszed to Table III presents the comparsons on dfferent tranng samples. The arrangement of the tranng and testng mages s the same wth the second experment. For each fxed

6 TABLE II COMPARISONS ON THE EXTENDED YALE B DATABASE ON DIFFERENCE TRAINING SAMPLES FOR EACH OBJECT. Algorthm Recognton Rate (Tranng Images=5) Recognton Rate (=10) Recognton Rate (=15) SR ± ± ± TASR ± ± ± Algorthm Recognton Rate (=20) Recognton Rate (=25) SR ± ± TASR ± ± TABLE III COMPARISONS ON THE CMU-PIE DATABASE ON DIFFERENCE TRAINING SAMPLES FOR EACH OBJECT. Algorthm Recognton Rate (Tranng Images=5) Recognton Rate (=10) Recognton Rate (=15) SR ± ± ± TASR ± ± ± tranng and testng mage, we perform 20 recognton tasks for each algorthm. From ths table, we see that our algorthm provdes the hghest recognton rate, especally when tranng mages become large. Furthermore, the varances of our method are lower than those of the orgnal SR, whch ndcates that our method s more stable under dfferent choces of tranng mages. V. CONCLUSION In ths paper, we propose a novel tone-aware sparse representaton model for face recognton. The man contrbuton of ths work s that we ncorporate tone varances nto sparse representaton model, makng t more effectve under unwell llumnaton varances caused by shadow, overexposure or others. Moreover, as a general sparse representaton model, the proposed TASR can be appled to other classfcatonbased applcatons, such as gender recognton and character recognton. REFERENCES [1] A. Beck and M. Teboulle. A fast teratve shrnkage-thresholdng algorthm for lnear nverse problems. SIAM Journal on Imagng Scences, 2(1): , [2] P. N. Belhumeur, J. P. Hespanha, and D. J. Kregman. Egenfaces vs. fsherfaces: Recognton usng class specfc lnear projecton. IEEE Transactons on Pattern Analyss and Machne Intellgence, 19(7): , [3] H. Chen, H. Chang, and T. Lu. Local dscrmnant embeddng and ts varants. In IEEE Conference on Computer Vson and Pattern Recognton, pages , [4] D. L. Donoho and Y. Tsag. Fast soluton of l 1 -norm mnmzaton problems when the soluton may be sparse. IEEE Transactons on Informaton Theory, 54(11): , [5] B. Efron, T. Haste, I. Johnstone, and R. Tbshran. Least angle regresson. Annals of Statstcs, 32: , [6] K. He, J. Sun, and X. Tang. Guded mage flterng. In European Conference on Computer Vson, pages 1 14, [7] R. He, W.-S. Zheng, and B.-G. Hu. Maxmum correntropy crteron for robust face recognton. IEEE Transactons on Pattern Analyss and Machne Intellgence, 33(8): , [8] X. He, S. Yan, Y. Hu, P. Nyog, and H. J. Zhang. Face Recognton Usng Laplacanfaces. IEEE Transactons on Pattern Analyss and Machne Intellgence, 27(3): , [9] J. P. Jones and L. A. Palmer. An evaluaton of the two-dmensonal Gabor flter model of smple receptve felds n cat strate cortex. Journal of neurophysology, 58(6): , [10] C. Kang, S. Lao, S. Xang, and C. Pan. kernel sparse representaton wth local patterns for face recognton. In IEEE Internatonal Conference on Image Processng, pages , [11] K.-C. Lee, J. Ho, and D. J. Kregman. Acqurng lnear subspaces for face recognton under varable lghtng. IEEE Transactons on Pattern Analyss and Machne Intellgence, 27(5): , [12] Z. Le, T. Ahonen, M. Petkänen, and S. Z. L. Local frequency descrptor for low-resoluton face recognton. In IEEE Internatonal Conference on Automatc Face and Gesture Recognton, pages , [13] A. Levn, D. Lschnsk, and Y. Wess. A closed-form soluton to natural mage mattng. IEEE Transactons on Pattern Analyss and Machne Intellgence, 30(2): , [14] S. Z. L and A. K. Jan, edtors. Handbook of Face Recognton, 2nd Edton. Sprnger, [15] A. Martnez and R. Benavente. The ar face database. Techncal Report 24, CVC, June [16] T. Ojala, M. Petkanen, and T. Maenpaa. Multresoluton gray-scale and rotaton nvarant texture classfcaton wth local bnary patterns. IEEE Transacton on Pattern Analyss and Machne Intellgence, 24(7): , [17] T. Sm, S. Baker, and M. Bsat. The cmu pose, llumnaton, and expresson (pe) database. In IEEE Internatonal Conference on Automatc Face and Gesture Recognton, pages 53 58, [18] R. Tbshran. Regresson shrnkage and selecton va the lasso. Journal of the Royal Statstcal Socety, 58(1): , [19] M. Turk and A. Pentland. Egenfaces for Recognton. Journal of Cogntve Neuroscence, 3(1):71 86, [20] J. Wrght, A. Y. Yang, A. Ganesh, S. S. Sastry, and Y. Ma. Robust face recognton va sparse representaton. IEEE Transactons on Pattern Analyss and Machne Intellgence, 31(2): , [21] S. Xe, S. Shan, X. Chen, and J. Chen. Fusng local patterns of gabor magntude and phase for face recognton. IEEE Transactons on Image Processng, 19(5): , [22] J. Yang and J. Yang. Why can lda be performed n pca transformed space. Pattern Recognton, 36: , [23] M. Yang and L. Zhang. Gabor feature based sparse representaton for face recognton wth gabor occluson dctonary. In European Conference on Computer Vson, pages , [24] M. Yang, L. Zhang, J. Yang, and D. Zhang. Robust sparse codng for face recognton. In IEEE Conference on Computer Vson and Pattern Recognton, pages , [25] X. Yuan and S. Yan. Vsual classfcaton wth mult-task jont sparse representaton. In IEEE Computer Socety Conference on Computer Vson and Pattern Recognton, pages , [26] W. Zhao, R. Chellappa, A. Rosenfeld, and P. J. Phllps. Face Recognton: A Lterature Survey. ACM Computng Surveys, 35(4): , 2003.

Discriminative Dictionary Learning with Pairwise Constraints

Discriminative 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 information

A Binarization Algorithm specialized on Document Images and Photos

A 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 information

Learning a Class-Specific Dictionary for Facial Expression Recognition

Learning a Class-Specific Dictionary for Facial Expression Recognition BULGARIAN ACADEMY OF SCIENCES CYBERNETICS AND INFORMATION TECHNOLOGIES Volume 16, No 4 Sofa 016 Prnt ISSN: 1311-970; Onlne ISSN: 1314-4081 DOI: 10.1515/cat-016-0067 Learnng a Class-Specfc Dctonary for

More information

Face Recognition University at Buffalo CSE666 Lecture Slides Resources:

Face 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 information

Support Vector Machines

Support 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 information

Computer Aided Drafting, Design and Manufacturing Volume 25, Number 2, June 2015, Page 14

Computer Aided Drafting, Design and Manufacturing Volume 25, Number 2, June 2015, Page 14 Computer Aded Draftng, Desgn and Manufacturng Volume 5, Number, June 015, Page 14 CADDM Face Recognton Algorthm Fusng Monogenc Bnary Codng and Collaboratve Representaton FU Yu-xan, PENG Lang-yu College

More information

EYE CENTER LOCALIZATION ON A FACIAL IMAGE BASED ON MULTI-BLOCK LOCAL BINARY PATTERNS

EYE CENTER LOCALIZATION ON A FACIAL IMAGE BASED ON MULTI-BLOCK LOCAL BINARY PATTERNS P.G. Demdov Yaroslavl State Unversty Anatoly Ntn, Vladmr Khryashchev, Olga Stepanova, Igor Kostern EYE CENTER LOCALIZATION ON A FACIAL IMAGE BASED ON MULTI-BLOCK LOCAL BINARY PATTERNS Yaroslavl, 2015 Eye

More information

Face Recognition by Fusing Binary Edge Feature and Second-order Mutual Information

Face Recognition by Fusing Binary Edge Feature and Second-order Mutual Information Face Recognton by Fusng Bnary Edge Feature and Second-order Mutual Informaton Jatao Song, Bejng Chen, We Wang, Xaobo Ren School of Electronc and Informaton Engneerng, Nngbo Unversty of Technology Nngbo,

More information

MULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION

MULTISPECTRAL 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 information

Local Quaternary Patterns and Feature Local Quaternary Patterns

Local Quaternary Patterns and Feature Local Quaternary Patterns Local Quaternary Patterns and Feature Local Quaternary Patterns Jayu Gu and Chengjun Lu The Department of Computer Scence, New Jersey Insttute of Technology, Newark, NJ 0102, USA Abstract - Ths paper presents

More information

Learning the Kernel Parameters in Kernel Minimum Distance Classifier

Learning 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 information

Detection of an Object by using Principal Component Analysis

Detection 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 information

Robust Kernel Representation with Statistical Local Features. for Face Recognition

Robust Kernel Representation with Statistical Local Features. for Face Recognition Robust Kernel Representaton wth Statstcal Local Features for Face Recognton Meng Yang, Student Member, IEEE, Le Zhang 1, Member, IEEE Smon C. K. Shu, Member, IEEE, and Davd Zhang, Fellow, IEEE Dept. of

More information

Face Recognition Based on SVM and 2DPCA

Face 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 information

Kernel Collaborative Representation Classification Based on Adaptive Dictionary Learning

Kernel 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 information

An Image Fusion Approach Based on Segmentation Region

An Image Fusion Approach Based on Segmentation Region Rong Wang, L-Qun Gao, Shu Yang, Yu-Hua Cha, and Yan-Chun Lu An Image Fuson Approach Based On Segmentaton Regon An Image Fuson Approach Based on Segmentaton Regon Rong Wang, L-Qun Gao, Shu Yang 3, Yu-Hua

More information

Robust Dictionary Learning with Capped l 1 -Norm

Robust Dictionary Learning with Capped l 1 -Norm Proceedngs of the Twenty-Fourth Internatonal Jont Conference on Artfcal Intellgence (IJCAI 205) Robust Dctonary Learnng wth Capped l -Norm Wenhao Jang, Fepng Ne, Heng Huang Unversty of Texas at Arlngton

More information

Subspace clustering. Clustering. Fundamental to all clustering techniques is the choice of distance measure between data points;

Subspace 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 information

General Regression and Representation Model for Face Recognition

General Regression and Representation Model for Face Recognition 013 IEEE Conference on Computer Vson and Pattern Recognton Workshops General Regresson and Representaton Model for Face Recognton Janjun Qan, Jan Yang School of Computer Scence and Engneerng Nanjng Unversty

More information

Recognizing Faces. Outline

Recognizing 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 information

Tsinghua University at TAC 2009: Summarizing Multi-documents by Information Distance

Tsinghua University at TAC 2009: Summarizing Multi-documents by Information Distance Tsnghua Unversty at TAC 2009: Summarzng Mult-documents by Informaton Dstance Chong Long, Mnle Huang, Xaoyan Zhu State Key Laboratory of Intellgent Technology and Systems, Tsnghua Natonal Laboratory for

More information

Feature Reduction and Selection

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 information

Robust Low-Rank Regularized Regression for Face Recognition with Occlusion

Robust Low-Rank Regularized Regression for Face Recognition with Occlusion Robust Low-Rank Regularzed Regresson for ace Recognton wth Occluson Janjun Qan, Jan Yang, anlong Zhang and Zhouchen Ln School of Computer Scence and ngneerng, Nanjng Unversty of Scence and echnology Key

More information

Content Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers

Content 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 information

TN348: Openlab Module - Colocalization

TN348: Openlab Module - Colocalization TN348: Openlab Module - Colocalzaton Topc The Colocalzaton module provdes the faclty to vsualze and quantfy colocalzaton between pars of mages. The Colocalzaton wndow contans a prevew of the two mages

More information

Classifier Selection Based on Data Complexity Measures *

Classifier 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 information

A Novel Adaptive Descriptor Algorithm for Ternary Pattern Textures

A Novel Adaptive Descriptor Algorithm for Ternary Pattern Textures A Novel Adaptve Descrptor Algorthm for Ternary Pattern Textures Fahuan Hu 1,2, Guopng Lu 1 *, Zengwen Dong 1 1.School of Mechancal & Electrcal Engneerng, Nanchang Unversty, Nanchang, 330031, Chna; 2. School

More information

A Fast Visual Tracking Algorithm Based on Circle Pixels Matching

A Fast Visual Tracking Algorithm Based on Circle Pixels Matching A Fast Vsual Trackng Algorthm Based on Crcle Pxels Matchng Zhqang Hou hou_zhq@sohu.com Chongzhao Han czhan@mal.xjtu.edu.cn Ln Zheng Abstract: A fast vsual trackng algorthm based on crcle pxels matchng

More information

LECTURE : MANIFOLD LEARNING

LECTURE : 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 information

Cluster Analysis of Electrical Behavior

Cluster 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 information

On Modeling Variations For Face Authentication

On Modeling Variations For Face Authentication On Modelng Varatons For Face Authentcaton Xaomng Lu Tsuhan Chen B.V.K. Vjaya Kumar Department of Electrcal and Computer Engneerng, Carnege Mellon Unversty Abstract In ths paper, we present a scheme for

More information

Fuzzy Filtering Algorithms for Image Processing: Performance Evaluation of Various Approaches

Fuzzy Filtering Algorithms for Image Processing: Performance Evaluation of Various Approaches Proceedngs of the Internatonal Conference on Cognton and Recognton Fuzzy Flterng Algorthms for Image Processng: Performance Evaluaton of Varous Approaches Rajoo Pandey and Umesh Ghanekar Department of

More information

Skew Angle Estimation and Correction of Hand Written, Textual and Large areas of Non-Textual Document Images: A Novel Approach

Skew Angle Estimation and Correction of Hand Written, Textual and Large areas of Non-Textual Document Images: A Novel Approach Angle Estmaton and Correcton of Hand Wrtten, Textual and Large areas of Non-Textual Document Images: A Novel Approach D.R.Ramesh Babu Pyush M Kumat Mahesh D Dhannawat PES Insttute of Technology Research

More information

Feature Extraction Based on Maximum Nearest Subspace Margin Criterion

Feature 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 information

Face Detection with Deep Learning

Face Detection with Deep Learning Face Detecton wth Deep Learnng Yu Shen Yus122@ucsd.edu A13227146 Kuan-We Chen kuc010@ucsd.edu A99045121 Yzhou Hao y3hao@ucsd.edu A98017773 Mn Hsuan Wu mhwu@ucsd.edu A92424998 Abstract The project here

More information

12/2/2009. Announcements. Parametric / Non-parametric. Case-Based Reasoning. Nearest-Neighbor on Images. Nearest-Neighbor Classification

12/2/2009. 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 information

Classification of Face Images Based on Gender using Dimensionality Reduction Techniques and SVM

Classification 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 information

Collaboratively Regularized Nearest Points for Set Based Recognition

Collaboratively 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 information

Competitive Sparse Representation Classification for Face Recognition

Competitive 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 information

Outline. Discriminative classifiers for image recognition. Where in the World? A nearest neighbor recognition example 4/14/2011. CS 376 Lecture 22 1

Outline. 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 information

A Computer Vision System for Automated Container Code Recognition

A Computer Vision System for Automated Container Code Recognition A Computer Vson System for Automated Contaner Code Recognton Hsn-Chen Chen, Chh-Ka Chen, Fu-Yu Hsu, Yu-San Ln, Yu-Te Wu, Yung-Nen Sun * Abstract Contaner code examnaton s an essental step n the contaner

More information

Gender Classification using Interlaced Derivative Patterns

Gender 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 information

Support Vector Machines

Support 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 information

FEATURE EXTRACTION. Dr. K.Vijayarekha. Associate Dean School of Electrical and Electronics Engineering SASTRA University, Thanjavur

FEATURE 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 information

Parallelism for Nested Loops with Non-uniform and Flow Dependences

Parallelism 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 information

Neurocomputing 101 (2013) Contents lists available at SciVerse ScienceDirect. Neurocomputing

Neurocomputing 101 (2013) Contents lists available at SciVerse ScienceDirect. Neurocomputing Neurocomputng (23) 4 5 Contents lsts avalable at ScVerse ScenceDrect Neurocomputng journal homepage: www.elsever.com/locate/neucom Localty constraned representaton based classfcaton wth spatal pyramd patches

More information

Edge Detection in Noisy Images Using the Support Vector Machines

Edge Detection in Noisy Images Using the Support Vector Machines Edge Detecton n Nosy Images Usng the Support Vector Machnes Hlaro Gómez-Moreno, Saturnno Maldonado-Bascón, Francsco López-Ferreras Sgnal Theory and Communcatons Department. Unversty of Alcalá Crta. Madrd-Barcelona

More information

Joint Example-based Depth Map Super-Resolution

Joint Example-based Depth Map Super-Resolution Jont Example-based Depth Map Super-Resoluton Yanje L 1, Tanfan Xue,3, Lfeng Sun 1, Janzhuang Lu,3,4 1 Informaton Scence and Technology Department, Tsnghua Unversty, Bejng, Chna Department of Informaton

More information

Combination of Local Multiple Patterns and Exponential Discriminant Analysis for Facial Recognition

Combination of Local Multiple Patterns and Exponential Discriminant Analysis for Facial Recognition Sensors & ransducers 203 by IFSA http://.sensorsportal.com Combnaton of Local Multple Patterns and Exponental Dscrmnant Analyss for Facal Recognton, 2 Lfang Zhou, 2 Bn Fang, 3 Wesheng L, 3 Ldou Wang College

More information

Analysis of Continuous Beams in General

Analysis of Continuous Beams in General Analyss of Contnuous Beams n General Contnuous beams consdered here are prsmatc, rgdly connected to each beam segment and supported at varous ponts along the beam. onts are selected at ponts of support,

More information

CS 534: Computer Vision Model Fitting

CS 534: Computer Vision Model Fitting CS 534: Computer Vson Model Fttng Sprng 004 Ahmed Elgammal Dept of Computer Scence CS 534 Model Fttng - 1 Outlnes Model fttng s mportant Least-squares fttng Maxmum lkelhood estmaton MAP estmaton Robust

More information

Multi-View Face Alignment Using 3D Shape Model for View Estimation

Multi-View Face Alignment Using 3D Shape Model for View Estimation Mult-Vew Face Algnment Usng 3D Shape Model for Vew Estmaton Yanchao Su 1, Hazhou A 1, Shhong Lao 1 Computer Scence and Technology Department, Tsnghua Unversty Core Technology Center, Omron Corporaton ahz@mal.tsnghua.edu.cn

More information

S1 Note. Basis functions.

S1 Note. Basis functions. S1 Note. Bass functons. Contents Types of bass functons...1 The Fourer bass...2 B-splne bass...3 Power and type I error rates wth dfferent numbers of bass functons...4 Table S1. Smulaton results of type

More information

A New Feature of Uniformity of Image Texture Directions Coinciding with the Human Eyes Perception 1

A New Feature of Uniformity of Image Texture Directions Coinciding with the Human Eyes Perception 1 A New Feature of Unformty of Image Texture Drectons Concdng wth the Human Eyes Percepton Xng-Jan He, De-Shuang Huang, Yue Zhang, Tat-Mng Lo 2, and Mchael R. Lyu 3 Intellgent Computng Lab, Insttute of Intellgent

More information

The Research of Support Vector Machine in Agricultural Data Classification

The 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 information

Human Face Recognition Using Generalized. Kernel Fisher Discriminant

Human Face Recognition Using Generalized. Kernel Fisher Discriminant Human Face Recognton Usng Generalzed Kernel Fsher Dscrmnant ng-yu Sun,2 De-Shuang Huang Ln Guo. Insttute of Intellgent Machnes, Chnese Academy of Scences, P.O.ox 30, Hefe, Anhu, Chna. 2. Department of

More information

Corner-Based Image Alignment using Pyramid Structure with Gradient Vector Similarity

Corner-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 information

Learning a Locality Preserving Subspace for Visual Recognition

Learning 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 information

A Probabilistic Approach to Detect Urban Regions from Remotely Sensed Images Based on Combination of Local Features

A Probabilistic Approach to Detect Urban Regions from Remotely Sensed Images Based on Combination of Local Features A Probablstc Approach to Detect Urban Regons from Remotely Sensed Images Based on Combnaton of Local Features Berl Sırmaçek German Aerospace Center (DLR) Remote Sensng Technology Insttute Weßlng, 82234,

More information

Palmprint Recognition Using Directional Representation and Compresses Sensing

Palmprint Recognition Using Directional Representation and Compresses Sensing Research Journal of Appled Scences, Engneerng and echnology 4(22): 4724-4728, 2012 ISSN: 2040-7467 Maxwell Scentfc Organzaton, 2012 Submtted: March 31, 2012 Accepted: Aprl 30, 2012 Publshed: November 15,

More information

High-Boost Mesh Filtering for 3-D Shape Enhancement

High-Boost Mesh Filtering for 3-D Shape Enhancement Hgh-Boost Mesh Flterng for 3-D Shape Enhancement Hrokazu Yagou Λ Alexander Belyaev y Damng We z Λ y z ; ; Shape Modelng Laboratory, Unversty of Azu, Azu-Wakamatsu 965-8580 Japan y Computer Graphcs Group,

More information

LOCAL FEATURE EXTRACTION AND MATCHING METHOD FOR REAL-TIME FACE RECOGNITION SYSTEM. Ho-Chul Shin, Hae Chul Choi and Seong-Dae Kim

LOCAL FEATURE EXTRACTION AND MATCHING METHOD FOR REAL-TIME FACE RECOGNITION SYSTEM. Ho-Chul Shin, Hae Chul Choi and Seong-Dae Kim LOCAL FEATURE EXTRACTIO AD MATCHIG METHOD FOR REAL-TIME FACE RECOGITIO SYSTEM Ho-Chul Shn, Hae Chul Cho and Seong-Dae Km Vsual Communcatons Lab., Department of EECS Korea Advanced Insttute of Scence and

More information

WIRELESS CAPSULE ENDOSCOPY IMAGE CLASSIFICATION BASED ON VECTOR SPARSE CODING.

WIRELESS CAPSULE ENDOSCOPY IMAGE CLASSIFICATION BASED ON VECTOR SPARSE CODING. WIRELESS CAPSULE ENDOSCOPY IMAGE CLASSIFICATION BASED ON VECTOR SPARSE CODING Tao Ma 1, Yuexan Zou 1 *, Zhqang Xang 1, Le L 1 and Y L 1 ADSPLAB/ELIP, School of ECE, Pekng Unversty, Shenzhen 518055, Chna

More information

An Application of the Dulmage-Mendelsohn Decomposition to Sparse Null Space Bases of Full Row Rank Matrices

An Application of the Dulmage-Mendelsohn Decomposition to Sparse Null Space Bases of Full Row Rank Matrices Internatonal Mathematcal Forum, Vol 7, 2012, no 52, 2549-2554 An Applcaton of the Dulmage-Mendelsohn Decomposton to Sparse Null Space Bases of Full Row Rank Matrces Mostafa Khorramzadeh Department of Mathematcal

More information

Scale Selective Extended Local Binary Pattern For Texture Classification

Scale Selective Extended Local Binary Pattern For Texture Classification Scale Selectve Extended Local Bnary Pattern For Texture Classfcaton Yutng Hu, Zhlng Long, and Ghassan AlRegb Multmeda & Sensors Lab (MSL) Georga Insttute of Technology 03/09/017 Outlne Texture Representaton

More information

NAG Fortran Library Chapter Introduction. G10 Smoothing in Statistics

NAG Fortran Library Chapter Introduction. G10 Smoothing in Statistics Introducton G10 NAG Fortran Lbrary Chapter Introducton G10 Smoothng n Statstcs Contents 1 Scope of the Chapter... 2 2 Background to the Problems... 2 2.1 Smoothng Methods... 2 2.2 Smoothng Splnes and Regresson

More information

Extraction of Texture Information from Fuzzy Run Length Matrix

Extraction of Texture Information from Fuzzy Run Length Matrix Internatonal Journal of Computer Applcatons (0975 8887) Volume 55 o.1, October 01 Extracton of Texture Informaton from Fuzzy Run Length Matrx Y. Venkateswarlu Head Dept. of CSE&IT Chatanya Insttuteof Engg.

More information

A PATTERN RECOGNITION APPROACH TO IMAGE SEGMENTATION

A 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 information

Determining the Optimal Bandwidth Based on Multi-criterion Fusion

Determining the Optimal Bandwidth Based on Multi-criterion Fusion Proceedngs of 01 4th Internatonal Conference on Machne Learnng and Computng IPCSIT vol. 5 (01) (01) IACSIT Press, Sngapore Determnng the Optmal Bandwdth Based on Mult-crteron Fuson Ha-L Lang 1+, Xan-Mn

More information

Two-Dimensional Supervised Discriminant Projection Method For Feature Extraction

Two-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 information

X- Chart Using ANOM Approach

X- Chart Using ANOM Approach ISSN 1684-8403 Journal of Statstcs Volume 17, 010, pp. 3-3 Abstract X- Chart Usng ANOM Approach Gullapall Chakravarth 1 and Chaluvad Venkateswara Rao Control lmts for ndvdual measurements (X) chart are

More information

Integrated Expression-Invariant Face Recognition with Constrained Optical Flow

Integrated 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 information

Infrared face recognition using texture descriptors

Infrared 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 information

Semi-Supervised Discriminant Analysis Based On Data Structure

Semi-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 information

Fast Feature Value Searching for Face Detection

Fast Feature Value Searching for Face Detection Vol., No. 2 Computer and Informaton Scence Fast Feature Value Searchng for Face Detecton Yunyang Yan Department of Computer Engneerng Huayn Insttute of Technology Hua an 22300, Chna E-mal: areyyyke@63.com

More information

Lobachevsky State University of Nizhni Novgorod. Polyhedron. Quick Start Guide

Lobachevsky State University of Nizhni Novgorod. Polyhedron. Quick Start Guide Lobachevsky State Unversty of Nzhn Novgorod Polyhedron Quck Start Gude Nzhn Novgorod 2016 Contents Specfcaton of Polyhedron software... 3 Theoretcal background... 4 1. Interface of Polyhedron... 6 1.1.

More information

SVM-based Learning for Multiple Model Estimation

SVM-based Learning for Multiple Model Estimation SVM-based Learnng for Multple Model Estmaton Vladmr Cherkassky and Yunqan Ma Department of Electrcal and Computer Engneerng Unversty of Mnnesota Mnneapols, MN 55455 {cherkass,myq}@ece.umn.edu Abstract:

More information

Smoothing Spline ANOVA for variable screening

Smoothing 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 information

A Robust Method for Estimating the Fundamental Matrix

A Robust Method for Estimating the Fundamental Matrix Proc. VIIth Dgtal Image Computng: Technques and Applcatons, Sun C., Talbot H., Ourseln S. and Adraansen T. (Eds.), 0- Dec. 003, Sydney A Robust Method for Estmatng the Fundamental Matrx C.L. Feng and Y.S.

More information

PERFORMANCE EVALUATION FOR SCENE MATCHING ALGORITHMS BY SVM

PERFORMANCE EVALUATION FOR SCENE MATCHING ALGORITHMS BY SVM PERFORMACE EVALUAIO FOR SCEE MACHIG ALGORIHMS BY SVM Zhaohu Yang a, b, *, Yngyng Chen a, Shaomng Zhang a a he Research Center of Remote Sensng and Geomatc, ongj Unversty, Shangha 200092, Chna - yzhac@63.com

More information

Efficient Sparsity Estimation via Marginal-Lasso Coding

Efficient Sparsity Estimation via Marginal-Lasso Coding Effcent Sparsty Estmaton va Margnal-Lasso Codng Tzu-Y Hung 1,JwenLu 2, Yap-Peng Tan 1, and Shenghua Gao 3 1 School of Electrcal and Electronc Engneerng, Nanyang Technologcal Unversty, Sngapore 2 Advanced

More information

On Incremental and Robust Subspace Learning

On Incremental and Robust Subspace Learning On Incremental and Robust Subspace Learnng Yongmn L, L-Qun Xu, Jason Morphett and Rchard Jacobs Content and Codng Lab, BT Exact pp1 MLB3/7, Oron Buldng, Adastral Park, Ipswch, IP5 3RE, UK Emal: Yongmn.L@bt.com

More information

Facial Expression Recognition Based on Local Binary Patterns and Local Fisher Discriminant Analysis

Facial 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 information

RECOGNIZING GENDER THROUGH FACIAL IMAGE USING SUPPORT VECTOR MACHINE

RECOGNIZING 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 information

Lecture 5: Multilayer Perceptrons

Lecture 5: Multilayer Perceptrons Lecture 5: Multlayer Perceptrons Roger Grosse 1 Introducton So far, we ve only talked about lnear models: lnear regresson and lnear bnary classfers. We noted that there are functons that can t be represented

More information

Laplacian Eigenmap for Image Retrieval

Laplacian 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 information

Combination of Color and Local Patterns as a Feature Vector for CBIR

Combination of Color and Local Patterns as a Feature Vector for CBIR Internatonal Journal of Computer Applcatons (975 8887) Volume 99 No.1, August 214 Combnaton of Color and Local Patterns as a Feature Vector for CBIR L.Koteswara Rao Asst.Professor, Dept of ECE Faculty

More information

Appearance-based Statistical Methods for Face Recognition

Appearance-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 information

3D vector computer graphics

3D vector computer graphics 3D vector computer graphcs Paolo Varagnolo: freelance engneer Padova Aprl 2016 Prvate Practce ----------------------------------- 1. Introducton Vector 3D model representaton n computer graphcs requres

More information

Histogram-Enhanced Principal Component Analysis for Face Recognition

Histogram-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 information

Research and Application of Fingerprint Recognition Based on MATLAB

Research and Application of Fingerprint Recognition Based on MATLAB Send Orders for Reprnts to reprnts@benthamscence.ae The Open Automaton and Control Systems Journal, 205, 7, 07-07 Open Access Research and Applcaton of Fngerprnt Recognton Based on MATLAB Nng Lu* Department

More information

PCA Based Gait Segmentation

PCA 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 information

Online Detection and Classification of Moving Objects Using Progressively Improving Detectors

Online Detection and Classification of Moving Objects Using Progressively Improving Detectors Onlne Detecton and Classfcaton of Movng Objects Usng Progressvely Improvng Detectors Omar Javed Saad Al Mubarak Shah Computer Vson Lab School of Computer Scence Unversty of Central Florda Orlando, FL 32816

More information

Machine Learning. Support Vector Machines. (contains material adapted from talks by Constantin F. Aliferis & Ioannis Tsamardinos, and Martin Law)

Machine 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 information

Histogram of Template for Pedestrian Detection

Histogram of Template for Pedestrian Detection PAPER IEICE TRANS. FUNDAMENTALS/COMMUN./ELECTRON./INF. & SYST., VOL. E85-A/B/C/D, No. xx JANUARY 20xx Hstogram of Template for Pedestran Detecton Shaopeng Tang, Non Member, Satosh Goto Fellow Summary In

More information

Outline. Type of Machine Learning. Examples of Application. Unsupervised Learning

Outline. 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 information

Palmprint Feature Extraction Using 2-D Gabor Filters

Palmprint Feature Extraction Using 2-D Gabor Filters Palmprnt Feature Extracton Usng 2-D Gabor Flters Wa Kn Kong Davd Zhang and Wenxn L Bometrcs Research Centre Department of Computng The Hong Kong Polytechnc Unversty Kowloon Hong Kong Correspondng author:

More information

Weighted Sparse Image Classification Based on Low Rank Representation

Weighted Sparse Image Classification Based on Low Rank Representation Copyrght 08 Tech Scence Press CMC, vol.56, no., pp.9-05, 08 Weghted Sparse Image Classfcaton Based on Low Rank Representaton Qd Wu, Ybng L, Yun Ln, * and Ruoln Zhou Abstract: The conventonal sparse representaton-based

More information

A Modified Median Filter for the Removal of Impulse Noise Based on the Support Vector Machines

A Modified Median Filter for the Removal of Impulse Noise Based on the Support Vector Machines A Modfed Medan Flter for the Removal of Impulse Nose Based on the Support Vector Machnes H. GOMEZ-MORENO, S. MALDONADO-BASCON, F. LOPEZ-FERRERAS, M. UTRILLA- MANSO AND P. GIL-JIMENEZ Departamento de Teoría

More information

Sum of Linear and Fractional Multiobjective Programming Problem under Fuzzy Rules Constraints

Sum 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 information