FACE detection and alignment are essential to many face

Size: px
Start display at page:

Download "FACE detection and alignment are essential to many face"

Transcription

1 IEEE SIGNAL PROCESSING LETTERS, VOL. 23, NO. 10, OCTOBER Jont Face Detecton and Algnment Usng Multtask Cascaded Convolutonal Networks Kapeng Zhang, Zhanpeng Zhang, Zhfeng L, Senor Member, IEEE, andyuqao, Senor Member, IEEE Abstract Face detecton and algnment n unconstraned envronment are challengng due to varous poses, llumnatons, and occlusons. Recent studes show that deep learnng approaches can acheve mpressve performance on these two tasks. In ths letter, we propose a deep cascaded multtask framework that explots the nherent correlaton between detecton and algnment to boost up ther performance. In partcular, our framework leverages a cascaded archtecture wth three stages of carefully desgned deep convolutonal networks to predct face and landmark locaton n a coarse-to-fne manner. In addton, we propose a new onlne hard sample mnng strategy that further mproves the performance n practce. Our method acheves superor accuracy over the stateof-the-art technques on the challengng face detecton dataset and benchmark and WIDER FACE benchmarks for face detecton, and annotated facal landmarks n the wld benchmark for face algnment, whle keeps real-tme performance. Index Terms Cascaded convolutonal neural network (CNN), face algnment, face detecton. I. INTRODUCTION FACE detecton and algnment are essental to many face applcatons, such as face recognton and facal expresson analyss. However, the large vsual varatons of faces, such as occlusons, large pose varatons, and extreme lghtngs, mpose great challenges for these tasks n real-world applcatons. The cascade face detector proposed by Vola and Jones [2] utlzes Haar-Lke features and AdaBoost to tran cascaded classfers, whch acheves good performance wth real-tme effcency. However, qute a few works [1], [3], [4] ndcate that ths knd of detector may degrade sgnfcantly n real-world applcatons wth larger vsual varatons of human faces even wth more advanced features and classfers. Besdes the cas- Manuscrpt receved Aprl 7, 2016; revsed June 12, 2016 and July 31, 2016; accepted August 10, Date of publcaton August 26, 2016; date of current verson September 9, Ths work was supported n part by External Cooperaton Program of BIC, n part by Chnese Academy of Scences (172644KYSB , KYSB ), n part by Shenzhen Research Program under Grant KQCX , Grant JSGG , Grant CXZZ , Grant CYJ , and Grant JCYJ , n part by Guangdong Research Program under Grant 2014B and Grant 2015B , n part by the Natural Scence Foundaton of Guangdong Provnce under Grant 2014A , and n part by the Key Laboratory of Human Machne Intellgence-Synergy Systems through the Chnese Academy of Scences. The assocate edtor coordnatng the revew of ths manuscrpt and approvng t for publcaton was Dr. Alexandre X. Falcao. K. Zhang, Z. L, and Y. Qao are wth Shenzhen Insttutes of Advanced Technology, Chnese Academy of Scences, Shenzhen , Chna (e-mal: kp.zhang@sat.ac.cn; zhfeng.l@sat.ac.cn; yu.qao@sat.ac.cn). Z. Zhang s wth the Department of Informaton Engneerng, The Chnese Unversty of Hong Kong, Hong Kong (e-mal: zz013@e.cuhk.edu.hk). Color versons of one or more of the fgures n ths letter are avalable onlne at Dgtal Object Identfer /LSP cade structure, Mathas et al. [5] [7] ntroduce deformable part models for face detecton and acheve remarkable performance. However, they are computatonally expensve and may usually requre expensve annotaton n the tranng stage. Recently, convolutonal neural networks (CNNs) acheve remarkable progresses n a varety of computer vson tasks, such as mage classfcaton [9] and face recognton [10]. Inspred by the sgnfcant successes of deep learnng methods n computer vson tasks, several studes utlze deep CNNs for face detecton. Yang et al. [11] tran deep CNNs for facal attrbute recognton to obtan hgh response n face regons, whch further yeld canddate wndows of faces. However, due to ts complex CNN structure, ths approach s tme costly n practce. L et al. [19] use cascaded CNNs for face detecton, but t requres boundng box calbraton from face detecton wth extra computatonal expense and gnores the nherent correlaton between facal landmarks localzaton and boundng box regresson. Face algnment also attracts extensve research nterests. Research works n ths area can be roughly dvded nto two categores, regresson-based methods [12], [13], [16], and template fttng approaches [7], [14], [15]. Recently, Zhang et al. [22] proposed to use facal attrbute recognton as an auxlary task to enhance face algnment performance usng deep CNN. However, most of prevous face detecton and face algnment methods gnore the nherent correlaton between these two tasks. Though several exstng works attempt to jontly solve them, there are stll lmtatons n these works. For example, Chen et al. [18] jontly conduct algnment and detecton wth random forest usng features of pxel value dfference. But, these handcraft features lmt ts performance a lot. Zhang et al. [20] use multtask CNN to mprove the accuracy of multvew face detecton, but the detecton recall s lmted by the ntal detecton wndow produced by a weak face detector. On the other hand, mnng hard samples n tranng s crtcal to strengthen the power of detector. However, tradtonal hard sample mnng usually performs n an offlne manner, whch sgnfcantly ncreases the manual operatons. It s desrable to desgn an onlne hard sample mnng method for face detecton, whch s adaptve to the current tranng status automatcally. In ths letter, we propose a new framework to ntegrate these two tasks usng unfed cascaded CNNs by multtask learnng. The proposed CNNs consst of three stages. In the frst stage, t produces canddate wndows quckly through a shallow CNN. Then, t refnes the wndows by rejectng a large number of nonfaces wndows through a more complex CNN. Fnally, t uses a more powerful CNN to refne the result agan and output fve facal landmarks postons. Thanks to ths multtask learnng framework, the performance of the algorthm can be notably mproved. The major contrbutons of ths letter are summarzed as follows: IEEE. Personal use s permtted, but republcaton/redstrbuton requres IEEE permsson. See standards/publcatons/rghts/ndex.html for more nformaton.

2 1500 IEEE SIGNAL PROCESSING LETTERS, VOL. 23, NO. 10, OCTOBER 2016 TABLE I COMPARISON OF SPEED AND VALIDATION ACCURACY OF OUR CNNS AND PREVIOUS CNNS [19] Group CNN 300 Forward Propagaton Valdaton Accuracy Group1 12-Net [19] s 94.4% P-Net s 94.6% Group2 24-Net [19] s 95.1% R-Net s 95.4% Group3 48-Net [19] s 93.2% O-Net s 95.4% Stage 3: Ths stage s smlar to the second stage, but n ths stage we am to dentfy face regons wth more supervson. In partcular, the network wll output fve facal landmarks postons. Fg. 1. Ppelne of our cascaded framework that ncludes three-stage multtask deep convolutonal networks. Frst, canddate wndows are produced through a fast P-Net. After that, we refne these canddates n the next stage through a R-Net. In the thrd stage, the O-Net produces fnal boundng box and facal landmarks poston. 1) We propose a new cascaded CNNs-based framework for jont face detecton and algnment, and carefully desgn lghtweght CNN archtecture for real-tme performance. 2) We propose an effectve method to conduct onlne hard sample mnng to mprove the performance. 3) Extensve experments are conducted on challengng benchmarks to show sgnfcant performance mprovement of the proposed approach compared to the state-of-the-art technques n both face detecton and face algnment tasks. II. APPROACH In ths secton, we wll descrbe our approach toward jont face detecton and algnment. A. Overall Framework The overall ppelne of our approach s shown n Fg. 1. Gven an mage, we ntally resze t to dfferent scales to buld an mage pyramd, whch s the nput of the followng three-stage cascaded framework. Stage 1: We explot a fully convolutonal network, called proposal network (P-Net), to obtan the canddate facal wndows and ther boundng box regresson vectors. Then canddates are calbrated based on the estmated boundng box regresson vectors. After that, we employ nonmaxmum suppresson (NMS) to merge hghly overlapped canddates. Stage 2: All canddates are fed to another CNN, called refne network (R-Net), whch further rejects a large number of false canddates, performs calbraton wth boundng box regresson, and conducts NMS. B. CNN Archtectures In [19], multple CNNs have been desgned for face detecton. However, we notce ts performance mght be lmted by the followng facts: 1) Some flters n convoluton layers lack dversty that may lmt ther dscrmnatve ablty; (2) compared to other multclass objecton detecton and classfcaton tasks, face detecton s a challengng bnary classfcaton task, so t may need less numbers of flters per layer. To ths end, we reduce the number of flters and change the 5 5 flter to 3 3 flter to reduce the computng, whle ncrease the depth to get better performance. Wth these mprovements, compared to the prevous archtecture n [19], we can get better performance wth less runtme (the results n tranng phase are shown n Table I. For far comparson, we use the same tranng and valdaton data n each group). Our CNN archtectures are shown n Fg. 2. We apply PReLU [30] as nonlnearty actvaton functon after the convoluton and fully connecton layers (except output layers). C. Tranng We leverage three tasks to tran our CNN detectors: face/nonface classfcaton, boundng box regresson, and facal landmark localzaton. 1) Face Classfcaton: The learnng objectve s formulated as a two-class classfcaton problem. For each sample x,we use the cross-entropy loss as L det = ( y det log (p )+ ( 1 y det ) (1 log(p )) ) (1) where p s the probablty produced by the network that ndcates sample x beng a face. The notaton y det {0, 1} denotes the ground-truth label. 2) Boundng Box Regresson: For each canddate wndow, we predct the offset between t and the nearest ground truth (.e., the boundng boxes left, top, heght, and wdth). The learnng objectve s formulated as a regresson problem, and we employ the Eucldean loss for each sample x L box = ŷbox y box 2 (2) 2 where ŷ box s the regresson target obtaned from the network and y box s the ground-truth coordnate. There are four coordnates, ncludng left top, heght and wdth, and thus y box R 4.

3 ZHANG et al.: JOINT FACE DETECTION AND ALIGNMENT USING MULTITASK CASCADED CONVOLUTIONAL NETWORKS 1501 Fg. 2. Archtectures of P-Net, R-Net, and O-Net, where MP means max poolng and Conv means convoluton. The step sze n convoluton and poolng s 1 and 2, respectvely. 3) Facal Landmark Localzaton: Smlar to boundng box regresson task, facal landmark detecton s formulated as a regresson problem and we mnmze the Eucldean loss as L landmark = ŷlandmark y landmark 2 (3) 2 where ŷ landmark s the facal landmark s coordnates obtaned from the network and y landmark s the ground-truth coordnate for the th sample. There are fve facal landmarks, ncludng left eye, rght eye, nose, left mouth corner, and rght mouth corner, and thus y landmark R 10. 4) Multsource Tranng: Snce we employ dfferent tasks n each CNN, there are dfferent types of tranng mages n the learnng process, such as face, nonface, and partally algned face. In ths case, some of the loss functons [.e., (1) (3)] are not used. For example, for the sample of background regon, we only compute L det, and the other two losses are set as 0. Ths can be mplemented drectly wth a sample type ndcator. Then, the overall learnng target can be formulated as mn N =1 j {det,box,landmark} α j β j Lj (4) where N s the number of tranng samples and α j denotes on the task mportance. We use (α det =1,α box = 0.5,α landmark = 0.5) n P-Net and R-Net, whle (α det = 1,α box = 0.5,α landmark =1) n output network (O-Net) for more accurate facal landmarks localzaton. β j {0, 1} s the sample type ndcator. In ths case, t s natural to employ stochastc gradent descent to tran these CNNs. 5) Onlne Hard Sample Mnng: Dfferent from conductng tradtonal hard sample mnng after orgnal classfer had been traned, we conduct onlne hard sample mnng n face/nonface classfcaton task whch s adaptve to the tranng process. In partcular, n each mnbatch, we sort the losses computed n the forward propagaton from all samples and select the top 70% of them as hard samples. Then, we only compute the gradents from these hard samples n the backward propagaton. That means we gnore the easy samples that are less helpful to strengthen the detector durng tranng. Experments show that ths strategy yelds better performance wthout manual sample selecton. Its effectveness s demonstrated n Secton III. Fg. 3. (a) Detecton performance of P-Net wth and wthout onlne hard sample mnng. (b) JA denotes jont face algnment learnng n O-Net whle No JA denotes do not jont t. No JA n BBR denotes use No JA O-Net for boundng box regresson. III. EXPERIMENTS In ths secton, we frst evaluate the effectveness of the proposed hard sample mnng strategy. Then, we compare our face detector and algnment aganst the state-of-the-art methods n face detecton dataset and benchmark (FDDB) [25], WIDER FACE [24], and annotated facal landmarks n the wld (AFLW) benchmark [8]. FDDB dataset contans the annotatons for 5171 faces n a set of 2845 mages. WIDER FACE dataset conssts of labeled face boundng boxes n mages, where 50% of them for testng (dvded nto three subsets accordng to the dffculty of mages), 40% for tranng, and the remanng for valdaton. AFLW contans the facal landmarks annotatons for faces and we use the same test subset as [22]. Fnally, we evaluate the computatonal effcency of our face detector. A. Tranng Data Snce we jontly perform face detecton and algnment, here we use followng four dfferent knds of data annotaton n our tranng process: 1) negatves: regons whose the ntersecton-over-unon (IoU) rato s less than 0.3 to any ground-truth faces; 2) postves: IoU above 0.65 to a ground truth face; 3) part faces: IoU between 0.4 and 0.65 to a ground truth face; and 4) landmark faces: faces labeled fve landmarks postons. There s an unclear gap between part faces and negatves, and there are varances among dfferent face annotatons. So, we choose IoU gap between 0.3 and 0.4. Negatves and postves are used for face classfcaton tasks, postves and part faces are

4 1502 IEEE SIGNAL PROCESSING LETTERS, VOL. 23, NO. 10, OCTOBER 2016 FDDB. Fg. 3(a) shows the results from two dfferent P-Nets on FDDB. It s clear that the onlne hard sample mnng s benefcal to mprove performance. It can brng about 1.5% overall performance mprovement on FDDB. C. Effectveness of Jont Detecton and Algnment To evaluate the contrbuton of jont detecton and algnment, we evaluate the performances of two dfferent O-Nets (jont facal landmarks regresson learnng and do not jont t) on FDDB (wth the same P-Net and R-Net). We also compare the performance of boundng box regresson n these two O- Nets. Fg. 3(b) suggests that jont landmark localzaton task learnng help to enhance both face classfcaton and boundng box regresson tasks. Fg. 4. (a) Evaluaton on FDDB. (b) (d) Evaluaton on three subsets of WIDER FACE. The number followng the method ndcates the average accuracy. Fg. 5. Evaluaton on AFLW for face algnment. TABLE II SPEED COMPARISON OF OUR METHOD AND OTHER METHODS Method GPU Speed Ours Nvda Ttan Black 99 FPS Cascade CNN [19] Nvda Ttan Black 100 FPS Faceness [11] Nvda Ttan Black 20 FPS DP2MFD [27] Nvda Tesla K FPS used for boundng box regresson, and landmark faces are used for facal landmark localzaton. Total tranng data are composed of 3:1:1:2 (negatves/postves/part face/landmark face) data. The tranng data collecton for each network s descrbed as follows: 1) P-Net: We randomly crop several patches from WIDER FACE [24] to collect postves, negatves, and part face. Then, we crop faces from CelebA [23] as landmark faces. 2) R-Net: We use the frst stage of our framework to detect faces from WIDER FACE [24] to collect postves, negatves, and part face whle landmark faces are detected from CelebA [23]. 3) O-Net: Smlar to R-Net to collect data, but we use the frst two stages of our framework to detect faces and collect data. B. Effectveness of Onlne Hard Sample Mnng To evaluate the contrbuton of the proposed onlne hard sample mnng strategy, we tran two P-Nets (wth and wthout onlne hard sample mnng) and compare ther performance on D. Evaluaton on Face Detecton To evaluate the performance of our face detecton method, we compare our method aganst the state-of-the-art methods [1], [5], [6], [11], [18], [19], [26] [29] n FDDB, and the state-of-the-art methods [1], [11], [24] n WIDER FACE. Fg. 4(a) (d) shows that our method consstently outperforms all the compared approaches by a large margn n both the benchmarks. We also evaluate our approach on some challengng photos. 1 E. Evaluaton on Face Algnment In ths part, we compare the face algnment performance of our method aganst the followng methods: RCPR [12], TSPM [7], Luxand face SDK [17], ESR [13], CDM [15], SDM [21], and TCDCN [22]. The mean error s measured by the dstances between the estmated landmarks and the ground truths, and normalzed wth respect to the nterocular dstance. Fg. 5 shows that our method outperforms all the state-of-the-art methods wth a margn. It also shows that our method shows less superorty n mouth corner localzaton. It may result from the small varances of expresson, whch has a sgnfcant nfluence n mouth corner poston, n our tranng data. F. Runtme Effcency Gven the cascade structure, our method can acheve hgh speed n jont face detecton and algnment. We compare our method wth the state-of-the-art technques on GPU and the results are shown n Table II. It s noted that our current mplementaton s based on unoptmzed MATLAB codes. IV. CONCLUSION In ths letter, we have proposed a multtask cascaded CNNs-based framework for jont face detecton and algnment. Expermental results demonstrated that our methods consstently outperform the state-of-the-art methods across several challengng benchmarks (ncludng FDDB and WIDER FACE benchmarks for face detecton, and AFLW benchmark for face algnment) whle acheves real-tme performance for VGA mages wth mnmum face sze. The three man contrbutons for performance mprovement are carefully desgned cascaded CNNs archtecture, onlne hard sample mnng strategy, and jont face algnment learnng. 1 Examples are shown n

5 ZHANG et al.: JOINT FACE DETECTION AND ALIGNMENT USING MULTITASK CASCADED CONVOLUTIONAL NETWORKS 1503 REFERENCES [1] B. Yang, J. Yan, Z. Le, and S. Z. L, Aggregate channel features for mult-vew face detecton, n IEEE Int. Jont Conf. Bometrcs, 2014, pp [2] P. Vola and M. J. Jones, Robust real-tme face detecton, Int. J. Comput. Vs., vol. 57, no. 2, pp , [3] M. T. Pham, Y. Gao, V. D. D. Hoang, and T. J. Cham, Fast polygonal ntegraton and ts applcaton n extendng Haar-lke features to mprove object detecton, n IEEE Conf. Comput. Vs. Pattern Recognt., 2010, pp [4] Q. Zhu, M. C. Yeh, K. T. Cheng, and S. Avdan, Fast human detecton usng a cascade of hstograms of orented gradents, n IEEE Comput. Conf. Comput. Vs. Pattern Recognt., 2006, pp [5] M. Mathas, R. Benenson, M. Pedersol, and L. Van Gool, Face detecton wthout bells and whstles, n Eur. Conf. Comput Vs.,2014,pp [6] J. Yan, Z. Le, L. Wen, and S. L, The fastest deformable part model for object detecton, n IEEE Conf. Comput. Vs. Pattern Recognt., 2014, pp [7] X. Zhu and D. Ramanan, Face detecton, pose estmaton, and landmark localzaton n the wld, n IEEE Conf. Comput. Vs. Pattern Recognt., 2012, pp [8] M. Köstnger, P. Wohlhart, P. M. Roth, and H. Bschof, Annotated facal landmarks n the wld: A large-scale, real-world database for facal landmark localzaton, n IEEE Conf. Comput. Vs. Pattern Recognt. Workshops, 2011, pp [9] A. Krzhevsky, I. Sutskever, and G. E. Hnton, ImageNet classfcaton wth deep convolutonal neural networks, n Adv. Neural Inf. Process. Syst., 2012, pp [10] Y. Sun, Y. Chen, X. Wang, and X. Tang, Deep learnng face representaton by jont dentfcaton-verfcaton, n Adv. Neural Inf. Process. Syst., 2014, pp [11] S. Yang, P. Luo, C. C. Loy, and X. Tang, From facal parts responses to face detecton: A deep learnng approach, n IEEE Int. Conf. Comput. Vs., 2015, pp [12] X. P. Burgos-Artzzu, P. Perona, and P. Dollar, Robust face landmark estmaton under occluson, n IEEE Int. Conf. Comput. Vs., 2013, pp [13] X. Cao, Y. We, F. Wen, and J. Sun, Face algnment by explct shape regresson, Int. J. Comput. Vs., vol. 107, no. 2, pp , [14] T. F. Cootes, G. J. Edwards, and C. J. Taylor, Actve appearance models, IEEE Trans. Pattern Anal. Mach. Intell., vol. 23, no. 6, pp , Jun [15] X. Yu, J. Huang, S. Zhang, W. Yan, and D. Metaxas, Pose-free facal landmark fttng va optmzed part mxtures and cascaded deformable shape model, n IEEE Int. Conf. Comput. Vs., 2013, pp [16] J. Zhang, S. Shan, M. Kan, and X. Chen, Coarse-to-fne auto-encoder networks (CFAN) for real-tme face algnment, n Eur. Conf. Comput. Vs., 2014, pp [17] Luxand Incorporated: Luxand face SDK. [Onlne]. Avalable: [18] D. Chen, S. Ren, Y. We, X. Cao, and J. Sun, Jont cascade face detecton and algnment, n Eur. Conf. Comput. Vs., 2014, pp [19] H. L, Z. Ln, X. Shen, J. Brandt, and G. Hua, A convolutonal neural network cascade for face detecton, n IEEE Conf. Comput. Vs. Pattern Recognt., 2015, pp [20] C. Zhang and Z. Zhang, Improvng multvew face detecton wth multtask deep convolutonal neural networks, n IEEE Wnter Conf. Appl. Comput. Vs., 2014, pp [21] X. Xong and F. Torre, Supervsed descent method and ts applcatons to face algnment, n IEEE Conf. Comput. Vs. Pattern Recognt., 2013, pp [22] Z. Zhang, P. Luo, C. C. Loy, and X. Tang, Facal landmark detecton by deep mult-task learnng, n Eur. Conf. Comput. Vs., 2014, pp [23] Z. Lu, P. Luo, X. Wang, and X. Tang, Deep learnng face attrbutes n the wld, n IEEE Int. Conf. Comput. Vs., 2015, pp [24] S. Yang, P. Luo, C. C. Loy, and X. Tang, WIDER FACE: A Face detecton benchmark, arxv: [25] V. Jan and E. G. Learned-Mller, FDDB: A benchmark for face detecton n unconstraned settngs, Unv. Massachusetts, Amherst, MA, USA, Tech. Rep. UMCS , [26] B. Yang, J. Yan, Z. Le, and S. Z. L, Convolutonal channel features, n IEEE Int. Conf. Comput. Vs., 2015, pp [27] R. Ranjan, V. M. Patel, and R. Chellappa, A deep pyramd deformable part model for face detecton, n IEEE Int. Conf. Bometrcs Theory, Appl. Syst., 2015, pp [28] G. Ghas and C. C. Fowlkes, Occluson coherence: Detectng and localzng occluded faces, arxv: [29] S. S. Farfade, M. J. Saberan, and L. J. L, Mult-vew face detecton usng deep convolutonal neural networks, n ACM Int. Conf. Multmeda Retreval, 2015, pp [30] K. He, X. Zhang, S. Ren, and J. Sun, Delvng deep nto rectfers: Surpassng human-level performance on ImageNet classfcaton, n IEEE Int. Conf. Comput. Vs., 2015, pp

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

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

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

Improvement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration

Improvement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration Improvement of Spatal Resoluton Usng BlockMatchng Based Moton Estmaton and Frame Integraton Danya Suga and Takayuk Hamamoto Graduate School of Engneerng, Tokyo Unversty of Scence, 6-3-1, Nuku, Katsuska-ku,

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

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

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

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

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

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

A Gradient Difference based Technique for Video Text Detection

A Gradient Difference based Technique for Video Text Detection A Gradent Dfference based Technque for Vdeo Text Detecton Palaahnakote Shvakumara, Trung Quy Phan and Chew Lm Tan School of Computng, Natonal Unversty of Sngapore {shva, phanquyt, tancl }@comp.nus.edu.sg

More information

Learning-based License Plate Detection on Edge Features

Learning-based License Plate Detection on Edge Features Learnng-based Lcense Plate Detecton on Edge Features Wng Teng Ho, Woo Hen Yap, Yong Haur Tay Computer Vson and Intellgent Systems (CVIS) Group Unverst Tunku Abdul Rahman, Malaysa wngteng_h@yahoo.com, woohen@yahoo.com,

More information

A Gradient Difference based Technique for Video Text Detection

A Gradient Difference based Technique for Video Text Detection 2009 10th Internatonal Conference on Document Analyss and Recognton A Gradent Dfference based Technque for Vdeo Text Detecton Palaahnakote Shvakumara, Trung Quy Phan and Chew Lm Tan School of Computng,

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

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

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

Improved Face Detection and Alignment using Cascade Deep Convolutional Network

Improved Face Detection and Alignment using Cascade Deep Convolutional Network Improved Face Detection and Alignment using Cascade Deep Convolutional Network Weilin Cong, Sanyuan Zhao, Hui Tian, and Jianbing Shen Beijing Key Laboratory of Intelligent Information Technology, School

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

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

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

An Efficient Face Detection Method Using Adaboost and Facial Parts

An Efficient Face Detection Method Using Adaboost and Facial Parts An Effcent Face Detecton Method Usng Adaboost and Facal Parts Yasaman Heydarzadeh, Abolfazl Torogh Haghghat Computer, IT and Electronc department Azad Unversty of Qazvn Tehran, Iran heydarzadeh@ qau.ac.r,

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

What is Object Detection? Face Detection using AdaBoost. Detection as Classification. Principle of Boosting (Schapire 90)

What is Object Detection? Face Detection using AdaBoost. Detection as Classification. Principle of Boosting (Schapire 90) CIS 5543 Coputer Vson Object Detecton What s Object Detecton? Locate an object n an nput age Habn Lng Extensons Vola & Jones, 2004 Dalal & Trggs, 2005 one or ultple objects Object segentaton Object detecton

More information

IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 25, NO. 4, APRIL

IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 25, NO. 4, APRIL IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 25, NO. 4, APRIL 2016 1713 Weakly Supervsed Fne-Graned Categorzaton Wth Part-Based Image Representaton Yu Zhang, Xu-Shen We, Janxn Wu, Member, IEEE, Janfe Ca,

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

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

SLAM Summer School 2006 Practical 2: SLAM using Monocular Vision

SLAM 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 information

Optimizing Document Scoring for Query Retrieval

Optimizing Document Scoring for Query Retrieval Optmzng Document Scorng for Query Retreval Brent Ellwen baellwe@cs.stanford.edu Abstract The goal of ths project was to automate the process of tunng a document query engne. Specfcally, I used machne learnng

More information

Deep Spatial-Temporal Joint Feature Representation for Video Object Detection

Deep Spatial-Temporal Joint Feature Representation for Video Object Detection sensors Artcle Deep Spatal-Temporal Jont Feature Representaton for Vdeo Object Detecton Baojun Zhao 1,2, Boya Zhao 1,2 ID, Lnbo Tang 1,2, *, Yuq Han 1,2 and Wenzheng Wang 1,2 1 School of Informaton and

More information

A Fast Content-Based Multimedia Retrieval Technique Using Compressed Data

A Fast Content-Based Multimedia Retrieval Technique Using Compressed Data A Fast Content-Based Multmeda Retreval Technque Usng Compressed Data Borko Furht and Pornvt Saksobhavvat NSF Multmeda Laboratory Florda Atlantc Unversty, Boca Raton, Florda 3343 ABSTRACT In ths paper,

More information

Simulation Based Analysis of FAST TCP using OMNET++

Simulation Based Analysis of FAST TCP using OMNET++ Smulaton Based Analyss of FAST TCP usng OMNET++ Umar ul Hassan 04030038@lums.edu.pk Md Term Report CS678 Topcs n Internet Research Sprng, 2006 Introducton Internet traffc s doublng roughly every 3 months

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

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

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

IEEE 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 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 information

Load Balancing for Hex-Cell Interconnection Network

Load Balancing for Hex-Cell Interconnection Network Int. J. Communcatons, Network and System Scences,,, - Publshed Onlne Aprl n ScRes. http://www.scrp.org/journal/jcns http://dx.do.org/./jcns.. Load Balancng for Hex-Cell Interconnecton Network Saher Manaseer,

More information

Focal Loss in 3D Object Detection

Focal Loss in 3D Object Detection 1 Focal Loss n 3D Object Detecton eng Yun1 Le Ta2 Yuan Wang2 Chengju Lu3 Mng Lu2 Fg. 1. Upper two rows show projected 3D object detecton results from the detector traned wth bnary cross entropy. Lower

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

Enhanced Face Detection Technique Based on Color Correction Approach and SMQT Features

Enhanced Face Detection Technique Based on Color Correction Approach and SMQT Features Journal of Software Engneerng and Applcatons, 2013, 6, 519-525 http://dx.do.org/10.4236/jsea.2013.610062 Publshed Onlne October 2013 (http://www.scrp.org/journal/jsea) 519 Enhanced Face Detecton Technque

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

A Unified Framework for Semantics and Feature Based Relevance Feedback in Image Retrieval Systems

A Unified Framework for Semantics and Feature Based Relevance Feedback in Image Retrieval Systems A Unfed Framework for Semantcs and Feature Based Relevance Feedback n Image Retreval Systems Ye Lu *, Chunhu Hu 2, Xngquan Zhu 3*, HongJang Zhang 2, Qang Yang * School of Computng Scence Smon Fraser Unversty

More information

Classifying Acoustic Transient Signals Using Artificial Intelligence

Classifying Acoustic Transient Signals Using Artificial Intelligence Classfyng Acoustc Transent Sgnals Usng Artfcal Intellgence Steve Sutton, Unversty of North Carolna At Wlmngton (suttons@charter.net) Greg Huff, Unversty of North Carolna At Wlmngton (jgh7476@uncwl.edu)

More information

Data Mining: Model Evaluation

Data Mining: Model Evaluation Data Mnng: Model Evaluaton Aprl 16, 2013 1 Issues: Evaluatng Classfcaton Methods Accurac classfer accurac: predctng class label predctor accurac: guessng value of predcted attrbutes Speed tme to construct

More information

An Optimal Algorithm for Prufer Codes *

An 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 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

arxiv: v2 [cs.cv] 9 Apr 2018

arxiv: v2 [cs.cv] 9 Apr 2018 Boundary-senstve Network for Portrat Segmentaton Xanzh Du 1, Xaolong Wang 2, Dawe L 2, Jngwen Zhu 2, Serafettn Tasc 2, Cameron Uprght 2, Stephen Walsh 2, Larry Davs 1 1 Computer Vson Lab, UMIACS, Unversty

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

Term Weighting Classification System Using the Chi-square Statistic for the Classification Subtask at NTCIR-6 Patent Retrieval Task

Term 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 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

Backpropagation: In Search of Performance Parameters

Backpropagation: In Search of Performance Parameters Bacpropagaton: In Search of Performance Parameters ANIL KUMAR ENUMULAPALLY, LINGGUO BU, and KHOSROW KAIKHAH, Ph.D. Computer Scence Department Texas State Unversty-San Marcos San Marcos, TX-78666 USA ae049@txstate.edu,

More information

Shape Representation Robust to the Sketching Order Using Distance Map and Direction Histogram

Shape Representation Robust to the Sketching Order Using Distance Map and Direction Histogram Shape Representaton Robust to the Sketchng Order Usng Dstance Map and Drecton Hstogram Department of Computer Scence Yonse Unversty Kwon Yun CONTENTS Revew Topc Proposed Method System Overvew Sketch Normalzaton

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

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

Switching Convolutional Neural Network for Crowd Counting

Switching Convolutional Neural Network for Crowd Counting Swtchng Convolutonal Neural Network for Crowd Countng Deepak Babu Sam Shv Surya R. Venkatesh Babu Indan Insttute of Scence Bangalore, INDIA 560012 bsdeepak@grads.cds.sc.ac.n, shv.surya314@gmal.com, venky@cds.sc.ac.n

More information

Real-time Joint Tracking of a Hand Manipulating an Object from RGB-D Input

Real-time Joint Tracking of a Hand Manipulating an Object from RGB-D Input Real-tme Jont Tracng of a Hand Manpulatng an Object from RGB-D Input Srnath Srdhar 1 Franzsa Mueller 1 Mchael Zollhöfer 1 Dan Casas 1 Antt Oulasvrta 2 Chrstan Theobalt 1 1 Max Planc Insttute for Informatcs

More information

Correlation Filters for Object Alignment

Correlation Filters for Object Alignment Correlaton Flters for Object Algnment Vshnu aresh Boddet Carnege Mellon Unversty naresh@cmuedu Takeo Kanade Carnege Mellon Unversty tk@cscmuedu BVK Vjaya Kumar Carnege Mellon Unversty kumar@ececmuedu Abstract

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

A Background Subtraction for a Vision-based User Interface *

A Background Subtraction for a Vision-based User Interface * A Background Subtracton for a Vson-based User Interface * Dongpyo Hong and Woontack Woo KJIST U-VR Lab. {dhon wwoo}@kjst.ac.kr Abstract In ths paper, we propose a robust and effcent background subtracton

More information

An Automatic Eye Detection Method for Gray Intensity Facial Images

An Automatic Eye Detection Method for Gray Intensity Facial Images www.ijcsi.org 272 An Automatc Eye Detecton Method for Gray Intensty Facal Images M. Hassaballah 1,2, Kenj Murakam 1, Shun Ido 1 1 Department of Computer Scence, Ehme Unversty, 790-8577, Japan 2 Department

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

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

Using Fuzzy Logic to Enhance the Large Size Remote Sensing Images

Using Fuzzy Logic to Enhance the Large Size Remote Sensing Images Internatonal Journal of Informaton and Electroncs Engneerng Vol. 5 No. 6 November 015 Usng Fuzzy Logc to Enhance the Large Sze Remote Sensng Images Trung Nguyen Tu Huy Ngo Hoang and Thoa Vu Van Abstract

More information

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

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

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

Audio Event Detection and classification using extended R-FCN Approach. Kaiwu Wang, Liping Yang, Bin Yang

Audio Event Detection and classification using extended R-FCN Approach. Kaiwu Wang, Liping Yang, Bin Yang Audo Event Detecton and classfcaton usng extended R-FCN Approach Kawu Wang, Lpng Yang, Bn Yang Key Laboratory of Optoelectronc Technology and Systems(Chongqng Unversty), Mnstry of Educaton, ChongQng Unversty,

More information

3D Face Reconstruction With Local Feature Refinement

3D Face Reconstruction With Local Feature Refinement ternatonal Journal of Multmeda and Ubqutous Engneerng Vol.9, No.8 (014), pp.59-7 http://dx.do.org/10.1457/jmue.014.9.8.06 3D Face Reconstructon Wth Local Feature Refnement Rudy Adpranata 1, Kartka Gunad

More information

Machine Learning 9. week

Machine 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 information

3D Face Reconstruction With Local Feature Refinement. Abstract

3D Face Reconstruction With Local Feature Refinement. Abstract , pp.6-74 http://dx.do.org/0.457/jmue.04.9.8.06 3D Face Reconstructon Wth Local Feature Refnement Rudy Adpranata, Kartka Gunad and Wendy Gunawan 3, formatcs Department, Petra Chrstan Unversty, Surabaya,

More information

Positive Semi-definite Programming Localization in Wireless Sensor Networks

Positive 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 information

Deformable Part-based Robust Face Detection under Occlusion by Using Face Decomposition into Face Components

Deformable Part-based Robust Face Detection under Occlusion by Using Face Decomposition into Face Components Deformable Part-based Robust Face Detecton under Occluson by Usng Face Decomposton nto Face Components Darjan Marčetć, Slobodan Rbarć Unversty of Zagreb, Faculty of Electrcal Engneerng and Computng, Croata

More information

Development of an Active Shape Model. Using the Discrete Cosine Transform

Development of an Active Shape Model. Using the Discrete Cosine Transform Development of an Actve Shape Model Usng the Dscrete Cosne Transform Kotaro Yasuda A Thess n The Department of Electrcal and Computer Engneerng Presented n Partal Fulfllment of the Requrements for the

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

Comparing Image Representations for Training a Convolutional Neural Network to Classify Gender

Comparing Image Representations for Training a Convolutional Neural Network to Classify Gender 2013 Frst Internatonal Conference on Artfcal Intellgence, Modellng & Smulaton Comparng Image Representatons for Tranng a Convolutonal Neural Network to Classfy Gender Choon-Boon Ng, Yong-Haur Tay, Bok-Mn

More information

arxiv: v1 [cs.cv] 6 Nov 2018

arxiv: v1 [cs.cv] 6 Nov 2018 Super-Identty Convolutonal Neural Network for Face Hallucnaton Kapeng Zhang 1, Zhanpeng Zhang 2, Cha-Wen Cheng 1,3, Wnston H. Hsu 1, Yu Qao 4, We Lu 5, and Tong Zhang 5 arxv:1811.02328v1 [cs.cv] 6 Nov

More information

Deep Classification in Large-scale Text Hierarchies

Deep Classification in Large-scale Text Hierarchies Deep Classfcaton n Large-scale Text Herarches Gu-Rong Xue Dkan Xng Qang Yang 2 Yong Yu Dept. of Computer Scence and Engneerng Shangha Jao-Tong Unversty {grxue, dkxng, yyu}@apex.sjtu.edu.cn 2 Hong Kong

More information

Problem Definitions and Evaluation Criteria for Computational Expensive Optimization

Problem 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 information

Journal of Chemical and Pharmaceutical Research, 2014, 6(6): Research Article. A selective ensemble classification method on microarray data

Journal of Chemical and Pharmaceutical Research, 2014, 6(6): Research Article. A selective ensemble classification method on microarray data Avalable onlne www.ocpr.com Journal of Chemcal and Pharmaceutcal Research, 2014, 6(6):2860-2866 Research Artcle ISSN : 0975-7384 CODEN(USA) : JCPRC5 A selectve ensemble classfcaton method on mcroarray

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

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

arxiv: v2 [cs.cv] 3 Aug 2017

arxiv: v2 [cs.cv] 3 Aug 2017 Swtchng Convolutonal Neural Network for Crowd Countng Deepak Babu Sam Shv Surya R. Venkatesh Babu Indan Insttute of Scence Bangalore, INDIA 560012 arxv:1708.00199v2 [cs.cv] 3 Aug 2017 bsdeepak@grads.cds.sc.ac.n,

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

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

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

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

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

Face Recognition via Centralized Coordinate Learning

Face Recognition via Centralized Coordinate Learning 1 Face Recognton va Centralzed Coordnate Learnng Xanbao Q, Le Zhang arxv:1801.05678v1 [cs.cv] 17 Jan 2018 Abstract Owe to the rapd development of deep neural network (DNN) technques and the emergence of

More information

Simulation: Solving Dynamic Models ABE 5646 Week 11 Chapter 2, Spring 2010

Simulation: Solving Dynamic Models ABE 5646 Week 11 Chapter 2, Spring 2010 Smulaton: Solvng Dynamc Models ABE 5646 Week Chapter 2, Sprng 200 Week Descrpton Readng Materal Mar 5- Mar 9 Evaluatng [Crop] Models Comparng a model wth data - Graphcal, errors - Measures of agreement

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

BOOSTING CLASSIFICATION ACCURACY WITH SAMPLES CHOSEN FROM A VALIDATION SET

BOOSTING CLASSIFICATION ACCURACY WITH SAMPLES CHOSEN FROM A VALIDATION SET 1 BOOSTING CLASSIFICATION ACCURACY WITH SAMPLES CHOSEN FROM A VALIDATION SET TZU-CHENG CHUANG School of Electrcal and Computer Engneerng, Purdue Unversty, West Lafayette, Indana 47907 SAUL B. GELFAND School

More information

Related-Mode Attacks on CTR Encryption Mode

Related-Mode Attacks on CTR Encryption Mode Internatonal Journal of Network Securty, Vol.4, No.3, PP.282 287, May 2007 282 Related-Mode Attacks on CTR Encrypton Mode Dayn Wang, Dongda Ln, and Wenlng Wu (Correspondng author: Dayn Wang) Key Laboratory

More information

2 ZHENG et al.: ASSOCIATING GROUPS OF PEOPLE (a) Ambgutes from person re dentfcaton n solaton (b) Assocatng groups of people may reduce ambgutes n mat

2 ZHENG et al.: ASSOCIATING GROUPS OF PEOPLE (a) Ambgutes from person re dentfcaton n solaton (b) Assocatng groups of people may reduce ambgutes n mat ZHENG et al.: ASSOCIATING GROUPS OF PEOPLE 1 Assocatng Groups of People We-Sh Zheng jason@dcs.qmul.ac.uk Shaogang Gong sgg@dcs.qmul.ac.uk Tao Xang txang@dcs.qmul.ac.uk School of EECS, Queen Mary Unversty

More information

Combined Object Detection and Segmentation

Combined Object Detection and Segmentation Combned Object Detecton and Segmentaton Jarch Vansteenberge, Masayuk Mukunok, and Mchhko Mnoh Abstract We develop a method for combned object detecton and segmentaton n natural scene. In our approach segmentaton

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

Virtual Machine Migration based on Trust Measurement of Computer Node

Virtual Machine Migration based on Trust Measurement of Computer Node Appled Mechancs and Materals Onlne: 2014-04-04 ISSN: 1662-7482, Vols. 536-537, pp 678-682 do:10.4028/www.scentfc.net/amm.536-537.678 2014 Trans Tech Publcatons, Swtzerland Vrtual Machne Mgraton based on

More information

Pose Estimation in Heavy Clutter using a Multi-Flash Camera

Pose Estimation in Heavy Clutter using a Multi-Flash Camera 2010 IEEE Internatonal Conference on Robotcs and Automaton Anchorage Conventon Dstrct May 3-8, 2010, Anchorage, Alaska, USA Pose Estmaton n Heavy Clutter usng a Mult-Flash Camera Mng-Yu Lu, Oncel Tuzel,

More information

Network Intrusion Detection Based on PSO-SVM

Network Intrusion Detection Based on PSO-SVM TELKOMNIKA Indonesan Journal of Electrcal Engneerng Vol.1, No., February 014, pp. 150 ~ 1508 DOI: http://dx.do.org/10.11591/telkomnka.v1.386 150 Network Intruson Detecton Based on PSO-SVM Changsheng Xang*

More information

Large-scale Web Video Event Classification by use of Fisher Vectors

Large-scale Web Video Event Classification by use of Fisher Vectors Large-scale Web Vdeo Event Classfcaton by use of Fsher Vectors Chen Sun and Ram Nevata Unversty of Southern Calforna, Insttute for Robotcs and Intellgent Systems Los Angeles, CA 90089, USA {chensun nevata}@usc.org

More information

Iris recognition algorithm based on point covering of high-dimensional space and neural network

Iris recognition algorithm based on point covering of high-dimensional space and neural network Irs recognton algorthm based on pont coverng of hgh-dmensonal space and neural network Wenmng Cao,, Janhu Hu, Gang Xao, Shoujue Wang The College of Informaton Engneerng, ZheJang Unversty of Technology,

More information

Improving Web Image Search using Meta Re-rankers

Improving Web Image Search using Meta Re-rankers VOLUME-1, ISSUE-V (Aug-Sep 2013) IS NOW AVAILABLE AT: www.dcst.com Improvng Web Image Search usng Meta Re-rankers B.Kavtha 1, N. Suata 2 1 Department of Computer Scence and Engneerng, Chtanya Bharath Insttute

More information