DOMAIN ADAPTATION BY ITERATIVE IMPROVEMENT OF SOFT-LABELING AND MAXIMIZATION OF NON-PARAMETRIC MUTUAL INFORMATION. M.N.A. Khan, Douglas R.

Size: px
Start display at page:

Download "DOMAIN ADAPTATION BY ITERATIVE IMPROVEMENT OF SOFT-LABELING AND MAXIMIZATION OF NON-PARAMETRIC MUTUAL INFORMATION. M.N.A. Khan, Douglas R."

Transcription

1 DOMAIN ADAPTATION BY ITERATIVE IMPROVEMENT OF SOFT-LABELING AND MAXIMIZATION OF NON-PARAMETRIC MUTUAL INFORMATION M.N.A. Khan, Douglas R. Heisterkamp Department of Computer Siene Oklahoma State University, Stillwater, OK {mohk, ABSTRACT Domain adaptation (DA) algorithms address the problem of distribution shift between training and testing data. Reent approahes transform data into a shared subspae by minimizing the shift between their marginal distributions. We propose a method to learn a ommon subspae that will leverage the lass onditional distributions of training samples along with reduing the marginal distribution shift. To learn the subspae, we employ a supervised tehnique based on nonparametri mutual information by induing soft label assignment for the unlabeled test data. The approah presents an iterative linear transformation for subspae learning by repeatedly updating test data preditions via soft-labeling and onsequently improving the subspae with maximization of mutual information. A set of omprehensive experiments on benhmark datasets is onduted to prove the effiay of our novel framework over state-of-the-art approahes. Index Terms Mutual information, soft-labeling, subspae, transfer learning. 1. INTRODUCTION To build a model for objet lass detetion or regression problem, it is generally assumed that training and testing data are sampled from the same distribution. This assumption is often hallenged in real life senario, i.e. the dataset on whih a model is trained (referred to as soure domain) may vary signifiantly from the test data distribution (target domain) (Figure 1). This may degrade test auray or performane of the trained model and entails the neessity of adapting that model suh that it an overome the distribution differene among training and testing datasets, widely known as dataset bias, domain shift or domain adaptation [1, 2, 3]. In this paper, we will refer the terms domain and data interhangeably. The ultimate goal is to ompensate the distribution divergene between soure and target domains suh that a lassifier trained using soure data an also perform well on diversely distributed but related target data. Formally, soure and target data are represented with same feature enodings, but their marginal probability distributions Domain Adaptation Fig. 1: Example of distribution differenes for the same biyle lass. DA methods will try to resolve this domain shift so that a lassifiation model an perform effetively aross domains. will be different. Here, both domains have the same set of lass labels. One pratial example is, learning an image lassifiation model with images generated from high-resolution amera whereas deploying that appliation into a devie with low-resolution amera. To deal with the domain shift problem, two different settings are usually onsidered: i) unsupervised domain adaptation [4, 5, 6, 7], where no labeled data available in target domain and ii) semi-supervised domain adaptation [8, 9], where only a few labeled data are available in target domain along with abundant labeled data of soure domain. In this paper, we will fous on the more hallenging unsupervised ase. One popular way to deal with the unsupervised ase is to find a ommon feature subspae that expresses shared strutures between soure and target data. For example, Fernando et al.[5] proposed a linear projetion funtion to align soure and target distributions. Some other approahes fous in instane re-weighting of soure data to math with target data in order to minimize their distribution differenes [6, 10, 11, 2]. Most of these works deal with orreting marginal distribution shift [5, 4, 6], ignoring the lass onditional distribution of soure data. Therefore, the learned subspae may not be optimal in terms of lass separation. This motivates us to utilize lass disriminative information of soure data to learn a ommon feature subspae with goals of resolving the distribution divergene and reating a disriminative subspae for unlabeled target data. A supervised tehnique based on maximization of nonparametri mutual information (MI) between data and orre-

2 sponding lass labels has been proved effetive in learning a disriminative subspae [12, 13]. Following prior work in domain adaptation setting with labeled soure and unlabeled target data [7], an iterative method is proposed. At eah iteration, a subspae is learned with MI maximization and then target data lass preditions are reated with soft-labeling (probability that a point belongs to a lass) by utilizing neighboring soure data in the learned subspae. The soft assignment of lass labels for target data is integrated into the objetive funtion [12] of MI maximization and influenes the next iteration subspae learning. These two steps ontinue till onverging to a final subspae. In summary, the ontributions of this work are i) utilizing lass label distributions of soure data along with all domain data distributions to develop a domain adaptation framework, ii) extending supervised method of MI to support unlabeled target data by induing soft-labeling and iii) proposing an iterative approah of ommon subspae learning based on maximization of non-parametri MI indued with soft-labeling. 2. SUBSPACE LEARNING BY MAXIMIZING SOFT-LABELING INDUCED QUADRATIC MUTUAL INFORMATION (QMI-S) Aording to information theoreti literature, Mutual Information (MI) is defined as a measure of dependene between random variables. Assume that X is a random variable representing d-dimensional data x R d and C is a disrete random variable representing lass labels {1, 2,..., N }, where N is the total number of lasses. Let p(x) be the density funtion of x and P () be the lass prior probability. Using Renyi s entropy, quadrati mutual information (QMI) is a non-parametri estimation of MI and is defined as [13], I(X, C) = (p(x, ) P ()p(x)) 2 dx x = p(x, ) 2 dx + P () 2 p(x) 2 dx x x 2 p(x, )P ()p(x)dx x = V in + V all 2V btw (1) Bouzas et al. proposed a subspae learning algorithm using trae-norm formulation of QMI [12]. The data distribution p(x) is estimated by a Parzen window method using a Gaussian kernel. A multivariate Gaussian N (x; µ, Σ) with mean vetor µ and ovariane matrix Σ is, N (x; µ, Σ) = 1 (2π)d Σ e( 1 2 (x µ)t Σ 1 (x µ)) A Parzen window density estimation of p(x) with IID samples x i is n 1 p(x) = n N ( x; x i, σ 2 I ) where n is the ardinality of dataset. We extend [12] by introduing a soft lass labeling into the QMI formulation. The lass prior probability P () an be expressed as n n P () = P ( x i )P (x i ) = P ( x i ) 1 n = S where P ( x i ) is the probability that x i belongs to lass (soft-labeling). For notational onveniene, we will further refer P () as S. The joint distribution p(x, ) an be expressed as p(x, ) = P ( x)p(x) = 1 n P ( x i )N ( x; x i, σ 2 I ) n Following [12] s approah, a trae-norm derivation using the above expressions for p(x), P () and p(x, ) is presented in Table 1 and allows Eq.(1) to be rewritten as where M = 1 n 2 ( I(X, C) = tr { Φ T MΦ } (2) z z T ) ( ) S 2 + n 2 11 T ( ) S 2 1 n 2 z T (3) where z = [P ( x 1 ), P ( x 2 ),..., P ( x n )] T R n 1 is the soft-labeling introdued into QMI and Φ R n m represents the mapped data points from the original feature spae to a kernel Hilbert spae using a mapping funtion ψ : X H, i.e., K = ΦΦ T. This formulation is referred to as QMI-S. A k-dimensional subspae is learned by finding a linear transformation W = Φ T A, A R n k, that maximizes QMI-S. That is, maximize I(X, C) = tr { W T Φ T MΦW } whih with unit ovariane onstraints beomes (see [12]) A tr { A T KM KA } = argmax A T KA=I tr{a T (4) KKA} with M = (M + M T )/2 (5) where M is a symmetri form of M. Centralized Gaussian kernel K R n n is defined as K = K g E n K g K g E n +E n K g E n, where K g (i, j) = N ( x i x j ; 0, 2σ 2 I ) and E n R n n onsists of elements eah equal to 1 n. The trae ratio problem of Eq.(4) an be approximated as a ratio trae optimization [14] and A is found using the generalized eigen value deomposition method [15]. After finding A, the data is projeted by X p = ΦW = KA.

3 Table 1: Derivation of V in, V all and V btw. Here K = ΦΦ T, z = [P ( x 1 ), P ( x 2 ),..., P ( x n )] T R n 1 and 1 = [1, 1,..., 1] T R n 1 V in = p(x, ) 2 dx V all = P () 2 p(x) 2 dx V btw = p(x, )P ()p(x)dx x x x n n n 2 P ( x i )P ( x j )N ( x i x j ; 0, 2σ 2 I ) = n n S 2 1 n 2 N ( x; x i, σ 2 I ) N ( x; x j, σ 2 I ) = n n ( ) 1 S n 2 P ( x i )N ( x i x j ; 0, 2σ 2 I ) j=1 j=1 i j ( ) n 2 z T Kz n 2 tr { T Kz z } n 2 S 2 1 T K1 n 2 S z T K1 ( ) { ( ) } n 2 S 2 tr { K11 T } n 2 tr { ( )} S K 1 z T { ( n 2 tr ( ) Φ T z z T Φ n 2 S 2 tr { Φ T (11 T )Φ } = 1 n 2 tr Φ T 1 ) } S z T Φ Construt kernel matrix K from input data X Step-I Compute M to find a low-dimensional subspae by QMI-S maximization Step-II Update label preditions for projeted target data onvergene Projeted soure and target data Fig. 2: Proposed method for iterative subspae learning based on QMI-S maximization. 3. ITERATIVE IMPROVEMENT OF SOFT-LABELING AND QMI-S SUBSPACE The data available are X R n d onsisting of soure domain data X s R ns d, target domain data X t R nt d and ground truth labels of soure data [y 1, y 2,..., y ns ] T, where n s and n t represent soure and target data size respetively and n = n s + n t. P ( x i ) is defined as follows. For x s i X s, labels are known so hard labeling an be used (i.e, P ( x s i ) = 1 if = y i else 0). On the other hand, target data are unlabeled, so a full distribution is used. For target data, P ( x t i ) is initialized with uniform label distribution i.e., for x t i X t, P ( x t i ) = 1 N for eah {1, 2,..., N }. If soure labels are from a lassifier instead of ground truth then the lassifier s P ( x s i ) an be used instead of hard labels. The proposed iterative QMI-S (Figure 2) onsists of, Step-I: M is omputed using Eq. (3) and (5). To find A for learning QMI-S subspae, Eq.(4) an be rewritten as KM KU = KKUΛ, where Λ is a diagonal matrix of eigen values and U is a matrix of orresponding eigen vetors. For omputational effiieny, we substitute KU with a new variable V i.e. KU = V, multiply both sides by K 1 and obtain M V = V Λ. This is a standard eigen problem where V and Λ represent matrix of eigen vetors and eigen values respetively. As M has rank N 1, the N 1 vetors with largest eigen values are seleted from V. Hene, A will be A = K 1 2 V and the projeted data X p R n k will be omputed as X p = KA = K 1 2 V. Step-II: Target data preditions P ( x t i ) are updated by applying a lassifier trained with projeted soure data. This will eventually update M of Step-I for the next iteration. P ( x s i ) will be remained same through out the iterations. The overall proess is summarized in Algorithm 1. Convergene riterion The proposed algorithm will Algorithm 1 Subspae learning based on iterative QMI-S. 1: Input: Data matrix X=[X s ;X t ] R n d where soure data X s R ns d and target data X t R nt d, soure data labels [y 1, y 2,..., y ns ] T. 2: Output: X p R n k, k-dimensional projeted data. 3: Initialization: For x t i X t, P ( x t i ) = 1 N for eah {1, 2,... N }. For x s i X s, P ( x s i ) = 1 if = y i and P ( x s i ) = 0 otherwise, for eah {1, 2,... N}. 4: Compute a entralized Gaussian kernel matrix, K R n n. 5: repeat Step-I: 6: Compute M matrix using Eq.(3) and (5). 7: Solve standard eigen problem, M V = V Λ. 8: Compute projeted data X p = K 1 2 V. Step-II: 9: Train a lassifier f using projeted soure data Xp s and apply it to update P ( x t i ) with soft-labeling. 10: until onvergene (defined in Setion 3). reah onvergene when subspae hange in two suessive iterations will be negligible. The subspae is defined by the basis vetors V. The differene beween two k-dimensional subspaes an be approximated as a subspae distane on a Grassmannian [16]. One suh distane metri measures the prinipal angle θ between V i and V i+1 of iterations i and i + 1 respetively [17, 16]. A onvergene threshold ɛ is set and the algorithm terminates when θ ɛ. At this state, lass preditions for target data are stabled. 4. DATASET AND EXPERIMENTS The proposed method is implemented and tested against popular benhmark datasets. Offie is a widely used image database for domain adaptation [18]. It ontains three different domains (Amazon, DSLR, Webam) of images aptured with varied settings and image onditions. Images of Amazon are downloaded from amazon site, DSLR ontains images aptured with high-resolution DSLR amera and Webam ontains images aptured with low-resolution web amera. Additionally, a popular dataset for objet reognition Calteh-256 [19] is used. The experiments will be onduted using these 4 domains with 10 ommon ategories seleted from eah of them (Bike, BakPak, Calulator, Headphone, Keyboard, Laptop, Monitor, Mouse, Mug, Projetor). From these 4 domains, a total of 12 DA sub-problems an be

4 Table 2: Comparative results in terms of lassifiation auray(%) of target data for 12 different sub-problems. Eah sub-problem onsists of soure target, where soure or target represents any of the four domains: C(Calteh-256), A(Amazon), W(Webam) and D(DSLR). Methods C A C W C D A C A W A D W C W A W D D C D A D W Avg Origfeat PCA GFK SA TCA TFL TJM QMI-H QMI-S reated, eah of whih ontains one soure and one target domain. Experimental setup The image representations published by Gong et al.[4] are used and the experimental protool of [6, 7] is followed. Input data are whitened with PCA preserving 95% of the data variane. Gaussian kernel σ is set to median of the pair-wise distanes of data in original feature spae. A K-nn lassifier with K set to log(n s ) + 1 heuristi [20] is used in line 9 of Algorithm 1. Convergene threshold ɛ = is used. The proposed QMI-S will be ompared with 7 other methods (see Table 2). They an be ategorized as follows, Without adaptation: Origfeat and PCA indiate the lassifiation auray of target data in original feature spae and PCA subspae respetively. Adaptation based on subspae alignment: It inludes geodesi flow kernel (GFK) [4], subspae alignment (SA) [5], transfer omponent analysis (TCA) [21] and transfer feature learning (TFL) [7]. Adaptation based on subspae alignment+instane reweighting: inludes transfer joint mathing (TJM) [6]. Analysis For eah of the 12 sub-problems with souretarget ombination (C A, C W et.), we reported the lassifiation auray(%) of target domain data. For all methods, auray is determined by a one nearest neighbor lassifier in the projeted spae or quoted from [6]. Our method shows improved performane ompared to others and outperforms them in 7 out of 12 sub-problems. In terms of average auray over all 12 ases, our method is 3.75% ahead of the losest average auray (TJM). We also onduted experiments with hard-labeling (denoted as QMI-H in Table 2) for target preditions. Iterative QMI-S approah learly shows its superiority as it onservatively updates target data labels, whereas QMI-H is aggressive and one a target point is falsely labeled, it is prone to stik with this label in suessive iterations. Figure 3 shows the affinity struture in the learned feature spae for three different methods inluding ours for the sub-problem C A. Projeted data are obtained using Origfeat, TJM and iterative QMI-S method. For illustration, 1051 projeted data from 5 different lasses are hosen Sr Trg Sr Trg (a) (b) () Fig. 3: Assesing the quality of feature subspae by similarity matrix for sub-problem C A using (a) original feature spae, (b) TJM and () iterative QMI-S. with the first set is 584 data from soure domain sorted by lass and the seond set is 467 data from target domain sorted by their predited labels. A similarity matrix with 25-nearest neighbors is onstruted (see Figure 3). The top-left and bottom-right sub-matries of eah sub-figure represent similarity inside a domain (soure or target). Iterative QMI-S exhibits more ompat blok diagonal struture (Figure 3()) ompared to TJM (Figure 3(b)). The top-right and bottomleft sub-matries represent similarity aross domains with better ompat blok diagonal struture generated by our method (whih represents better within-lass similarity). In terms of omputational ost, the average over all 12 sub-problems of the number of iterations of QMI-S until onvergene is 25. The dominate ost in eah iteration is solving an eigen value deomposition of a n n for a set of largest eigenvalues and eigenvetors. 5. CONCLUSION This paper proposes a domain adaptation algorithm based on soft-labeling indued quadrati mutual information. Unlike other subspae alignment methods, the goal is to utilize lass onditional distribution of soure domain to learn a ommon subspae with improved lass separation suh that a lassifier trained with projeted soure data an be applied to projeted target data effetively. In future work, we are planning to inorporate instane weighting into this framework in order to failitate subspae learning only with losely related data samples aross domains along with minimizing the impat of unrelated soure samples.

5 6. REFERENCES [1] Antonio Torralba, Alexei Efros, et al., Unbiased look at dataset bias, in IEEE Conferene on Computer Vision and Pattern Reognition, 2011, pp [2] Jiayuan Huang, Arthur Gretton, Karsten M Borgwardt, Bernhard Shölkopf, and Alex J Smola, Correting sample seletion bias by unlabeled data, in Pro. Advanes in Neural Information Proessing Systems, 2006, pp [3] Sinno Jialin Pan and Qiang Yang, A survey on transfer learning, IEEE Transations on Knowledge and Data Engineering, vol. 22, no. 10, pp , [4] Boqing Gong, Yuan Shi, Fei Sha, and Kristen Grauman, Geodesi flow kernel for unsupervised domain adaptation, in IEEE Conferene on Computer Vision and Pattern Reognition, 2012, 2012, pp [5] Basura Fernando, Amaury Habrard, Mar Sebban, and Tinne Tuytelaars, Unsupervised visual domain adaptation using subspae alignment, in IEEE International Conferene on Computer Vision, 2013, 2013, pp [6] Mingsheng Long, Jianmin Wang, Guiguang Ding, Jiaguang Sun, and Philip S Yu, Transfer joint mathing for unsupervised domain adaptation, in IEEE Conferene on Computer Vision and Pattern Reognition, 2014, pp [7] Mingsheng Long, Jianmin Wang, Guiguang Ding, Jiaguang Sun, and Philip S Yu, Transfer feature learning with joint distribution adaptation, in IEEE International Conferene on Computer Vision, IEEE, 2013, pp [8] Jeff Donahue, Judy Hoffman, Erid Rodner, Kate Saenko, and Trevor Darrell, Semi-supervised domain adaptation with instane onstraints, in IEEE Conferene on Computer Vision and Pattern Reognition, 2013, 2013, pp [9] Hal Daumé III, Abhishek Kumar, and Avishek Saha, Frustratingly easy semi-supervised domain adaptation, in Pro. of the 2010 Workshop on Domain Adaptation for Natural Language Proessing, 2010, pp [11] Wen-Sheng Chu, Fernando De la Torre, and Jeffrey F Cohn, Seletive transfer mahine for personalized faial ation unit detetion, in IEEE Conferene on Computer Vision and Pattern Reognition, 2013, 2013, pp [12] Dimitrios Bouzas, Nikolaos Arvanitopoulos, and Anastasios Tefas, Graph embedded nonparametri mutual information for supervised dimensionality redution, IEEE Trans. on Neural Networks and Learning Systems, vol. 26, [13] Kari Torkkola, Feature extration by non parametri mutual information maximization, The Journal of Mahine Learning Researh, vol. 3, pp , [14] Yangqing Jia, Feiping Nie, and Changshui Zhang, Trae ratio problem revisited, IEEE Trans. on Neural Networks, vol. 20, no. 4, pp , [15] Keinosuke Fukunaga, Introdution to statistial pattern reognition, Aademi press, [16] Jayaraman J Thiagarajan and Karthikeyan Natesan Ramamurthy, Subspae learning using onsensus on the grassmannian manifold, in IEEE International Conferene on Aoustis, Speeh and Signal Proessing (ICASSP). IEEE, 2015, pp [17] Jihun Hamm and Daniel D Lee, Grassmann disriminant analysis: a unifying view on subspae-based learning, in Pro. 25th International Conferene on Mahine learning. ACM, 2008, pp [18] Kate Saenko, Brian Kulis, Mario Fritz, and Trevor Darrell, Adapting visual ategory models to new domains, in Pro. ECCV 2010, pp Springer, [19] G. Griffin, A. Holub, and P. Perona, Calteh-256 objet ategory dataset, Teh. Rep. 7694, California Institute of Tehnology, [20] Ulrike Von Luxburg, A tutorial on spetral lustering, Statistis and Computing, vol. 17, no. 4, pp , [21] Sinno Jialin Pan, Ivor W Tsang, James T Kwok, and Qiang Yang, Domain adaptation via transfer omponent analysis, IEEE Trans. on Neural Networks, vol. 22, no. 2, pp , [10] Minmin Chen, Kilian Q Weinberger, and John Blitzer, Co-training for domain adaptation, in Pro. Advanes in Neural Information Proessing Systems, 2011, pp

Adapting Instance Weights For Unsupervised Domain Adaptation Using Quadratic Mutual Information And Subspace Learning

Adapting Instance Weights For Unsupervised Domain Adaptation Using Quadratic Mutual Information And Subspace Learning Adapting Instane Weights For Unsupervised Domain Adaptation Using Quadrati Mutual Information And ubspae Learning M.N.A. Khan, Douglas R. Heisterkamp Department of Computer iene Oklahoma tate University,

More information

Learning Convention Propagation in BeerAdvocate Reviews from a etwork Perspective. Abstract

Learning Convention Propagation in BeerAdvocate Reviews from a etwork Perspective. Abstract CS 9 Projet Final Report: Learning Convention Propagation in BeerAdvoate Reviews from a etwork Perspetive Abstrat We look at the way onventions propagate between reviews on the BeerAdvoate dataset, and

More information

Cluster-Based Cumulative Ensembles

Cluster-Based Cumulative Ensembles Cluster-Based Cumulative Ensembles Hanan G. Ayad and Mohamed S. Kamel Pattern Analysis and Mahine Intelligene Lab, Eletrial and Computer Engineering, University of Waterloo, Waterloo, Ontario N2L 3G1,

More information

EASY TRANSFER LEARNING BY EXPLOITING INTRA-DOMAIN STRUCTURES

EASY TRANSFER LEARNING BY EXPLOITING INTRA-DOMAIN STRUCTURES EASY TRANSFER LEARNING BY EXPLOITING INTRA-DOMAIN STRUCTURES Jindong Wang 1, Yiqiang Chen 1,, Han Yu 2, Meiyu Huang 3, Qiang Yang 4 1 Beiing Key Lab. of Mobile Computing and Pervasive Devie, Inst. of Computing

More information

the data. Structured Principal Component Analysis (SPCA)

the data. Structured Principal Component Analysis (SPCA) Strutured Prinipal Component Analysis Kristin M. Branson and Sameer Agarwal Department of Computer Siene and Engineering University of California, San Diego La Jolla, CA 9193-114 Abstrat Many tasks involving

More information

The Minimum Redundancy Maximum Relevance Approach to Building Sparse Support Vector Machines

The Minimum Redundancy Maximum Relevance Approach to Building Sparse Support Vector Machines The Minimum Redundany Maximum Relevane Approah to Building Sparse Support Vetor Mahines Xiaoxing Yang, Ke Tang, and Xin Yao, Nature Inspired Computation and Appliations Laboratory (NICAL), Shool of Computer

More information

KERNEL SPARSE REPRESENTATION WITH LOCAL PATTERNS FOR FACE RECOGNITION

KERNEL SPARSE REPRESENTATION WITH LOCAL PATTERNS FOR FACE RECOGNITION KERNEL SPARSE REPRESENTATION WITH LOCAL PATTERNS FOR FACE RECOGNITION Cuiui Kang 1, Shengai Liao, Shiming Xiang 1, Chunhong Pan 1 1 National Laboratory of Pattern Reognition, Institute of Automation, Chinese

More information

A Novel Validity Index for Determination of the Optimal Number of Clusters

A Novel Validity Index for Determination of the Optimal Number of Clusters IEICE TRANS. INF. & SYST., VOL.E84 D, NO.2 FEBRUARY 2001 281 LETTER A Novel Validity Index for Determination of the Optimal Number of Clusters Do-Jong KIM, Yong-Woon PARK, and Dong-Jo PARK, Nonmembers

More information

arxiv: v1 [cs.cv] 6 Jul 2016

arxiv: v1 [cs.cv] 6 Jul 2016 arxiv:607.079v [cs.cv] 6 Jul 206 Deep CORAL: Correlation Alignment for Deep Domain Adaptation Baochen Sun and Kate Saenko University of Massachusetts Lowell, Boston University Abstract. Deep neural networks

More information

arxiv: v1 [cs.db] 13 Sep 2017

arxiv: v1 [cs.db] 13 Sep 2017 An effiient lustering algorithm from the measure of loal Gaussian distribution Yuan-Yen Tai (Dated: May 27, 2018) In this paper, I will introdue a fast and novel lustering algorithm based on Gaussian distribution

More information

Boosted Random Forest

Boosted Random Forest Boosted Random Forest Yohei Mishina, Masamitsu suhiya and Hironobu Fujiyoshi Department of Computer Siene, Chubu University, 1200 Matsumoto-ho, Kasugai, Aihi, Japan {mishi, mtdoll}@vision.s.hubu.a.jp,

More information

A Coarse-to-Fine Classification Scheme for Facial Expression Recognition

A Coarse-to-Fine Classification Scheme for Facial Expression Recognition A Coarse-to-Fine Classifiation Sheme for Faial Expression Reognition Xiaoyi Feng 1,, Abdenour Hadid 1 and Matti Pietikäinen 1 1 Mahine Vision Group Infoteh Oulu and Dept. of Eletrial and Information Engineering

More information

Improved Vehicle Classification in Long Traffic Video by Cooperating Tracker and Classifier Modules

Improved Vehicle Classification in Long Traffic Video by Cooperating Tracker and Classifier Modules Improved Vehile Classifiation in Long Traffi Video by Cooperating Traker and Classifier Modules Brendan Morris and Mohan Trivedi University of California, San Diego San Diego, CA 92093 {b1morris, trivedi}@usd.edu

More information

Rotation Invariant Spherical Harmonic Representation of 3D Shape Descriptors

Rotation Invariant Spherical Harmonic Representation of 3D Shape Descriptors Eurographis Symposium on Geometry Proessing (003) L. Kobbelt, P. Shröder, H. Hoppe (Editors) Rotation Invariant Spherial Harmoni Representation of 3D Shape Desriptors Mihael Kazhdan, Thomas Funkhouser,

More information

Discrete sequential models and CRFs. 1 Case Study: Supervised Part-of-Speech Tagging

Discrete sequential models and CRFs. 1 Case Study: Supervised Part-of-Speech Tagging 0-708: Probabilisti Graphial Models 0-708, Spring 204 Disrete sequential models and CRFs Leturer: Eri P. Xing Sribes: Pankesh Bamotra, Xuanhong Li Case Study: Supervised Part-of-Speeh Tagging The supervised

More information

Learning Discriminative and Shareable Features. Scene Classificsion

Learning Discriminative and Shareable Features. Scene Classificsion Learning Disriminative and Shareable Features for Sene Classifiation Zhen Zuo, Gang Wang, Bing Shuai, Lifan Zhao, Qingxiong Yang, and Xudong Jiang Nanyang Tehnologial University, Singapore, Advaned Digital

More information

Dr.Hazeem Al-Khafaji Dept. of Computer Science, Thi-Qar University, College of Science, Iraq

Dr.Hazeem Al-Khafaji Dept. of Computer Science, Thi-Qar University, College of Science, Iraq Volume 4 Issue 6 June 014 ISSN: 77 18X International Journal of Advaned Researh in Computer Siene and Software Engineering Researh Paper Available online at: www.ijarsse.om Medial Image Compression using

More information

Geodesic Flow Kernel for Unsupervised Domain Adaptation

Geodesic Flow Kernel for Unsupervised Domain Adaptation Geodesic Flow Kernel for Unsupervised Domain Adaptation Boqing Gong University of Southern California Joint work with Yuan Shi, Fei Sha, and Kristen Grauman 1 Motivation TRAIN TEST Mismatch between different

More information

TUMOR DETECTION IN MRI BRAIN IMAGE SEGMENTATION USING PHASE CONGRUENCY MODIFIED FUZZY C MEAN ALGORITHM

TUMOR DETECTION IN MRI BRAIN IMAGE SEGMENTATION USING PHASE CONGRUENCY MODIFIED FUZZY C MEAN ALGORITHM TUMOR DETECTION IN MRI BRAIN IMAGE SEGMENTATION USING PHASE CONGRUENCY MODIFIED FUZZY C MEAN ALGORITHM M. Murugeswari 1, M.Gayathri 2 1 Assoiate Professor, 2 PG Sholar 1,2 K.L.N College of Information

More information

New Fuzzy Object Segmentation Algorithm for Video Sequences *

New Fuzzy Object Segmentation Algorithm for Video Sequences * JOURNAL OF INFORMATION SCIENCE AND ENGINEERING 24, 521-537 (2008) New Fuzzy Obet Segmentation Algorithm for Video Sequenes * KUO-LIANG CHUNG, SHIH-WEI YU, HSUEH-JU YEH, YONG-HUAI HUANG AND TA-JEN YAO Department

More information

Accepted Manuscript. Domain Class Consistency based Transfer Learning for Image Classification Across Domains. Lei Zhang, Jian Yang, David Zhang

Accepted Manuscript. Domain Class Consistency based Transfer Learning for Image Classification Across Domains. Lei Zhang, Jian Yang, David Zhang Aepted Manusript Domain lass onsisteny based ransfer Learning for Image lassifiation Aross Domains Lei Zhang Jian Yang David Zhang PII: 000-055(6)335-9 DOI: 0.06/j.ins.07.08.034 Referene: IN 3038 o appear

More information

Hyperspectral Images Classification Using Energy Profiles of Spatial and Spectral Features

Hyperspectral Images Classification Using Energy Profiles of Spatial and Spectral Features journal homepage: www.elsevier.om Hyperspetral Images Classifiation Using Energy Profiles of Spatial and Spetral Features Hamid Reza Shahdoosti a a Hamedan University of ehnology, Department of Eletrial

More information

Capturing Large Intra-class Variations of Biometric Data by Template Co-updating

Capturing Large Intra-class Variations of Biometric Data by Template Co-updating Capturing Large Intra-lass Variations of Biometri Data by Template Co-updating Ajita Rattani University of Cagliari Piazza d'armi, Cagliari, Italy ajita.rattani@diee.unia.it Gian Lua Marialis University

More information

Unsupervised Stereoscopic Video Object Segmentation Based on Active Contours and Retrainable Neural Networks

Unsupervised Stereoscopic Video Object Segmentation Based on Active Contours and Retrainable Neural Networks Unsupervised Stereosopi Video Objet Segmentation Based on Ative Contours and Retrainable Neural Networks KLIMIS NTALIANIS, ANASTASIOS DOULAMIS, and NIKOLAOS DOULAMIS National Tehnial University of Athens

More information

Time delay estimation of reverberant meeting speech: on the use of multichannel linear prediction

Time delay estimation of reverberant meeting speech: on the use of multichannel linear prediction University of Wollongong Researh Online Faulty of Informatis - apers (Arhive) Faulty of Engineering and Information Sienes 7 Time delay estimation of reverberant meeting speeh: on the use of multihannel

More information

Extracting Partition Statistics from Semistructured Data

Extracting Partition Statistics from Semistructured Data Extrating Partition Statistis from Semistrutured Data John N. Wilson Rihard Gourlay Robert Japp Mathias Neumüller Department of Computer and Information Sienes University of Strathlyde, Glasgow, UK {jnw,rsg,rpj,mathias}@is.strath.a.uk

More information

Evolutionary Feature Synthesis for Image Databases

Evolutionary Feature Synthesis for Image Databases Evolutionary Feature Synthesis for Image Databases Anlei Dong, Bir Bhanu, Yingqiang Lin Center for Researh in Intelligent Systems University of California, Riverside, California 92521, USA {adong, bhanu,

More information

Weak Dependence on Initialization in Mixture of Linear Regressions

Weak Dependence on Initialization in Mixture of Linear Regressions Proeedings of the International MultiConferene of Engineers and Computer Sientists 8 Vol I IMECS 8, Marh -6, 8, Hong Kong Weak Dependene on Initialization in Mixture of Linear Regressions Ryohei Nakano

More information

Spatial-Aware Collaborative Representation for Hyperspectral Remote Sensing Image Classification

Spatial-Aware Collaborative Representation for Hyperspectral Remote Sensing Image Classification Spatial-Aware Collaborative Representation for Hyperspetral Remote Sensing Image ifiation Junjun Jiang, Member, IEEE, Chen Chen, Member, IEEE, Yi Yu, Xinwei Jiang, and Jiayi Ma Member, IEEE Representation-residual

More information

2017 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media,

2017 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, 2017 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any urrent or future media, inluding reprinting/republishing this material for advertising

More information

Abstract. Key Words: Image Filters, Fuzzy Filters, Order Statistics Filters, Rank Ordered Mean Filters, Channel Noise. 1.

Abstract. Key Words: Image Filters, Fuzzy Filters, Order Statistics Filters, Rank Ordered Mean Filters, Channel Noise. 1. Fuzzy Weighted Rank Ordered Mean (FWROM) Filters for Mixed Noise Suppression from Images S. Meher, G. Panda, B. Majhi 3, M.R. Meher 4,,4 Department of Eletronis and I.E., National Institute of Tehnology,

More information

Performance of Histogram-Based Skin Colour Segmentation for Arms Detection in Human Motion Analysis Application

Performance of Histogram-Based Skin Colour Segmentation for Arms Detection in Human Motion Analysis Application World Aademy of Siene, Engineering and Tehnology 8 009 Performane of Histogram-Based Skin Colour Segmentation for Arms Detetion in Human Motion Analysis Appliation Rosalyn R. Porle, Ali Chekima, Farrah

More information

3-D IMAGE MODELS AND COMPRESSION - SYNTHETIC HYBRID OR NATURAL FIT?

3-D IMAGE MODELS AND COMPRESSION - SYNTHETIC HYBRID OR NATURAL FIT? 3-D IMAGE MODELS AND COMPRESSION - SYNTHETIC HYBRID OR NATURAL FIT? Bernd Girod, Peter Eisert, Marus Magnor, Ekehard Steinbah, Thomas Wiegand Te {girod eommuniations Laboratory, University of Erlangen-Nuremberg

More information

Plot-to-track correlation in A-SMGCS using the target images from a Surface Movement Radar

Plot-to-track correlation in A-SMGCS using the target images from a Surface Movement Radar Plot-to-trak orrelation in A-SMGCS using the target images from a Surfae Movement Radar G. Golino Radar & ehnology Division AMS, Italy ggolino@amsjv.it Abstrat he main topi of this paper is the formulation

More information

Exploiting Enriched Contextual Information for Mobile App Classification

Exploiting Enriched Contextual Information for Mobile App Classification Exploiting Enrihed Contextual Information for Mobile App Classifiation Hengshu Zhu 1 Huanhuan Cao 2 Enhong Chen 1 Hui Xiong 3 Jilei Tian 2 1 University of Siene and Tehnology of China 2 Nokia Researh Center

More information

Self-Adaptive Parent to Mean-Centric Recombination for Real-Parameter Optimization

Self-Adaptive Parent to Mean-Centric Recombination for Real-Parameter Optimization Self-Adaptive Parent to Mean-Centri Reombination for Real-Parameter Optimization Kalyanmoy Deb and Himanshu Jain Department of Mehanial Engineering Indian Institute of Tehnology Kanpur Kanpur, PIN 86 {deb,hjain}@iitk.a.in

More information

Video Data and Sonar Data: Real World Data Fusion Example

Video Data and Sonar Data: Real World Data Fusion Example 14th International Conferene on Information Fusion Chiago, Illinois, USA, July 5-8, 2011 Video Data and Sonar Data: Real World Data Fusion Example David W. Krout Applied Physis Lab dkrout@apl.washington.edu

More information

Graph-Based vs Depth-Based Data Representation for Multiview Images

Graph-Based vs Depth-Based Data Representation for Multiview Images Graph-Based vs Depth-Based Data Representation for Multiview Images Thomas Maugey, Antonio Ortega, Pasal Frossard Signal Proessing Laboratory (LTS), Eole Polytehnique Fédérale de Lausanne (EPFL) Email:

More information

A scheme for racquet sports video analysis with the combination of audio-visual information

A scheme for racquet sports video analysis with the combination of audio-visual information A sheme for raquet sports video analysis with the ombination of audio-visual information Liyuan Xing a*, Qixiang Ye b, Weigang Zhang, Qingming Huang a and Hua Yu a a Graduate Shool of the Chinese Aadamy

More information

A New RBFNDDA-KNN Network and Its Application to Medical Pattern Classification

A New RBFNDDA-KNN Network and Its Application to Medical Pattern Classification A New RBFNDDA-KNN Network and Its Appliation to Medial Pattern Classifiation Shing Chiang Tan 1*, Chee Peng Lim 2, Robert F. Harrison 3, R. Lee Kennedy 4 1 Faulty of Information Siene and Tehnology, Multimedia

More information

Fast Rigid Motion Segmentation via Incrementally-Complex Local Models

Fast Rigid Motion Segmentation via Incrementally-Complex Local Models Fast Rigid Motion Segmentation via Inrementally-Complex Loal Models Fernando Flores-Mangas Allan D. Jepson Department of Computer Siene, University of Toronto {mangas,jepson}@s.toronto.edu Abstrat The

More information

Detection and Recognition of Non-Occluded Objects using Signature Map

Detection and Recognition of Non-Occluded Objects using Signature Map 6th WSEAS International Conferene on CIRCUITS, SYSTEMS, ELECTRONICS,CONTROL & SIGNAL PROCESSING, Cairo, Egypt, De 9-31, 007 65 Detetion and Reognition of Non-Oluded Objets using Signature Map Sangbum Park,

More information

Using Augmented Measurements to Improve the Convergence of ICP

Using Augmented Measurements to Improve the Convergence of ICP Using Augmented Measurements to Improve the onvergene of IP Jaopo Serafin, Giorgio Grisetti Dept. of omputer, ontrol and Management Engineering, Sapienza University of Rome, Via Ariosto 25, I-0085, Rome,

More information

FUZZY WATERSHED FOR IMAGE SEGMENTATION

FUZZY WATERSHED FOR IMAGE SEGMENTATION FUZZY WATERSHED FOR IMAGE SEGMENTATION Ramón Moreno, Manuel Graña Computational Intelligene Group, Universidad del País Vaso, Spain http://www.ehu.es/winto; {ramon.moreno,manuel.grana}@ehu.es Abstrat The

More information

Shape Outlier Detection Using Pose Preserving Dynamic Shape Models

Shape Outlier Detection Using Pose Preserving Dynamic Shape Models Shape Outlier Detetion Using Pose Preserving Dynami Shape Models Chan-Su Lee Ahmed Elgammal Department of Computer Siene, Rutgers University, Pisataway, NJ 8854 USA CHANSU@CS.RUTGERS.EDU ELGAMMAL@CS.RUTGERS.EDU

More information

Approximate logic synthesis for error tolerant applications

Approximate logic synthesis for error tolerant applications Approximate logi synthesis for error tolerant appliations Doohul Shin and Sandeep K. Gupta Eletrial Engineering Department, University of Southern California, Los Angeles, CA 989 {doohuls, sandeep}@us.edu

More information

Face and Facial Feature Tracking for Natural Human-Computer Interface

Face and Facial Feature Tracking for Natural Human-Computer Interface Fae and Faial Feature Traking for Natural Human-Computer Interfae Vladimir Vezhnevets Graphis & Media Laboratory, Dept. of Applied Mathematis and Computer Siene of Mosow State University Mosow, Russia

More information

Polyhedron Volume-Ratio-based Classification for Image Recognition

Polyhedron Volume-Ratio-based Classification for Image Recognition Polyhedron Volume-Ratio-based Classifiation for Image Reognition Qingxiang Feng, Jeng-Shyang Pan, Senior Member, IEEE, Jar-Ferr Yang, Fellow, IEEE and Yang-ing Chou Abstrat In this paper, a novel method,

More information

Landmarks-based Kernelized Subspace Alignment for Unsupervised Domain Adaptation

Landmarks-based Kernelized Subspace Alignment for Unsupervised Domain Adaptation Landmarks-based Kernelized Subspace Alignment for Unsupervised Domain Adaptation Rahaf Aljundi, Re mi Emonet, Damien Muselet and Marc Sebban rahaf.aljundi@gmail.com remi.emonet, damien.muselet, marc.sebban

More information

Transition Detection Using Hilbert Transform and Texture Features

Transition Detection Using Hilbert Transform and Texture Features Amerian Journal of Signal Proessing 1, (): 35-4 DOI: 1.593/.asp.1.6 Transition Detetion Using Hilbert Transform and Texture Features G. G. Lashmi Priya *, S. Domni Department of Computer Appliations, National

More information

NONLINEAR BACK PROJECTION FOR TOMOGRAPHIC IMAGE RECONSTRUCTION. Ken Sauer and Charles A. Bouman

NONLINEAR BACK PROJECTION FOR TOMOGRAPHIC IMAGE RECONSTRUCTION. Ken Sauer and Charles A. Bouman NONLINEAR BACK PROJECTION FOR TOMOGRAPHIC IMAGE RECONSTRUCTION Ken Sauer and Charles A. Bouman Department of Eletrial Engineering, University of Notre Dame Notre Dame, IN 46556, (219) 631-6999 Shool of

More information

Incomplete Multi-Modal Visual Data Grouping

Incomplete Multi-Modal Visual Data Grouping Proeedings of the Twenty-ifth International Joint Conferene on Artifiial Intelligene (IJCAI-16) Inomplete Multi-Modal Visual Data Grouping Handong Zhao, Hongfu Liu and Yun u Department of Eletrial and

More information

A Partial Sorting Algorithm in Multi-Hop Wireless Sensor Networks

A Partial Sorting Algorithm in Multi-Hop Wireless Sensor Networks A Partial Sorting Algorithm in Multi-Hop Wireless Sensor Networks Abouberine Ould Cheikhna Department of Computer Siene University of Piardie Jules Verne 80039 Amiens Frane Ould.heikhna.abouberine @u-piardie.fr

More information

Semi-supervised feature extraction for population structure identification using the Laplacian linear discriminant Usman Roshan 1,* 1

Semi-supervised feature extraction for population structure identification using the Laplacian linear discriminant Usman Roshan 1,* 1 Population and Geneti Analysis Semi-supervised feature extration for population struture identifiation using the Laplaian linear disriminant Usman Roshan 1,* 1 Department of Computer Siene, New Jersey

More information

Gait Based Human Recognition with Various Classifiers Using Exhaustive Angle Calculations in Model Free Approach

Gait Based Human Recognition with Various Classifiers Using Exhaustive Angle Calculations in Model Free Approach Ciruits and Systems, 2016, 7, 1465-1475 Published Online June 2016 in SiRes. http://www.sirp.org/journal/s http://dx.doi.org/10.4236/s.2016.78128 Gait Based Human Reognition with Various Classifiers Using

More information

Automatic Physical Design Tuning: Workload as a Sequence Sanjay Agrawal Microsoft Research One Microsoft Way Redmond, WA, USA +1-(425)

Automatic Physical Design Tuning: Workload as a Sequence Sanjay Agrawal Microsoft Research One Microsoft Way Redmond, WA, USA +1-(425) Automati Physial Design Tuning: Workload as a Sequene Sanjay Agrawal Mirosoft Researh One Mirosoft Way Redmond, WA, USA +1-(425) 75-357 sagrawal@mirosoft.om Eri Chu * Computer Sienes Department University

More information

Smooth Trajectory Planning Along Bezier Curve for Mobile Robots with Velocity Constraints

Smooth Trajectory Planning Along Bezier Curve for Mobile Robots with Velocity Constraints Smooth Trajetory Planning Along Bezier Curve for Mobile Robots with Veloity Constraints Gil Jin Yang and Byoung Wook Choi Department of Eletrial and Information Engineering Seoul National University of

More information

Multi-Piece Mold Design Based on Linear Mixed-Integer Program Toward Guaranteed Optimality

Multi-Piece Mold Design Based on Linear Mixed-Integer Program Toward Guaranteed Optimality INTERNATIONAL CONFERENCE ON MANUFACTURING AUTOMATION (ICMA200) Multi-Piee Mold Design Based on Linear Mixed-Integer Program Toward Guaranteed Optimality Stephen Stoyan, Yong Chen* Epstein Department of

More information

One Against One or One Against All : Which One is Better for Handwriting Recognition with SVMs?

One Against One or One Against All : Which One is Better for Handwriting Recognition with SVMs? One Against One or One Against All : Whih One is Better for Handwriting Reognition with SVMs? Jonathan Milgram, Mohamed Cheriet, Robert Sabourin To ite this version: Jonathan Milgram, Mohamed Cheriet,

More information

Exploring the Commonality in Feature Modeling Notations

Exploring the Commonality in Feature Modeling Notations Exploring the Commonality in Feature Modeling Notations Miloslav ŠÍPKA Slovak University of Tehnology Faulty of Informatis and Information Tehnologies Ilkovičova 3, 842 16 Bratislava, Slovakia miloslav.sipka@gmail.om

More information

Volume 3, Issue 9, September 2013 International Journal of Advanced Research in Computer Science and Software Engineering

Volume 3, Issue 9, September 2013 International Journal of Advanced Research in Computer Science and Software Engineering Volume 3, Issue 9, September 2013 ISSN: 2277 128X International Journal of Advaned Researh in Computer Siene and Software Engineering Researh Paper Available online at: www.ijarsse.om A New-Fangled Algorithm

More information

Trajectory Tracking Control for A Wheeled Mobile Robot Using Fuzzy Logic Controller

Trajectory Tracking Control for A Wheeled Mobile Robot Using Fuzzy Logic Controller Trajetory Traking Control for A Wheeled Mobile Robot Using Fuzzy Logi Controller K N FARESS 1 M T EL HAGRY 1 A A EL KOSY 2 1 Eletronis researh institute, Cairo, Egypt 2 Faulty of Engineering, Cairo University,

More information

Detecting Outliers in High-Dimensional Datasets with Mixed Attributes

Detecting Outliers in High-Dimensional Datasets with Mixed Attributes Deteting Outliers in High-Dimensional Datasets with Mixed Attributes A. Koufakou, M. Georgiopoulos, and G.C. Anagnostopoulos 2 Shool of EECS, University of Central Florida, Orlando, FL, USA 2 Dept. of

More information

A Hybrid Neuro-Genetic Approach to Short-Term Traffic Volume Prediction

A Hybrid Neuro-Genetic Approach to Short-Term Traffic Volume Prediction A Hybrid Neuro-Geneti Approah to Short-Term Traffi Volume Predition 1. Introdution Shahriar Afandizadeh 1,*, Jalil Kianfar 2 Reeived: January 2003, Revised: July 2008, Aepted: January 2009 Abstrat: This

More information

Chromaticity-matched Superimposition of Foreground Objects in Different Environments

Chromaticity-matched Superimposition of Foreground Objects in Different Environments FCV216, the 22nd Korea-Japan Joint Workshop on Frontiers of Computer Vision Chromatiity-mathed Superimposition of Foreground Objets in Different Environments Yohei Ogura Graduate Shool of Siene and Tehnology

More information

A {k, n}-secret Sharing Scheme for Color Images

A {k, n}-secret Sharing Scheme for Color Images A {k, n}-seret Sharing Sheme for Color Images Rastislav Luka, Konstantinos N. Plataniotis, and Anastasios N. Venetsanopoulos The Edward S. Rogers Sr. Dept. of Eletrial and Computer Engineering, University

More information

Relevance for Computer Vision

Relevance for Computer Vision The Geometry of ROC Spae: Understanding Mahine Learning Metris through ROC Isometris, by Peter A. Flah International Conferene on Mahine Learning (ICML-23) http://www.s.bris.a.uk/publiations/papers/74.pdf

More information

An Efficient and Scalable Approach to CNN Queries in a Road Network

An Efficient and Scalable Approach to CNN Queries in a Road Network An Effiient and Salable Approah to CNN Queries in a Road Network Hyung-Ju Cho Chin-Wan Chung Dept. of Eletrial Engineering & Computer Siene Korea Advaned Institute of Siene and Tehnology 373- Kusong-dong,

More information

A Descriptive Framework for the Multidimensional Medical Data Mining and Representation

A Descriptive Framework for the Multidimensional Medical Data Mining and Representation Journal of Computer Siene 7 (4): 519-55, 011 ISSN 1549-3636 011 Siene Publiations A Desriptive Framework for the Multidimensional Medial Data Mining and Representation Veeramalai Sankaradass and Kannan

More information

Towards Optimal Naive Bayes Nearest Neighbor

Towards Optimal Naive Bayes Nearest Neighbor Towards Optimal Naive Bayes Nearest Neighbor Régis Behmo 1, Paul Marombes 1,2, Arnak Dalalyan 2,andVéronique Prinet 1 1 NLPR / LIAMA, Institute of Automation, Chinese Aademy of Sienes 2 IMAGINE, LIGM,

More information

Adaptation of Rigid Registration Algorithm to the Fingerprints Identification

Adaptation of Rigid Registration Algorithm to the Fingerprints Identification Amerian Journal of Software Engineering and Appliations 205; 4(6): 07-4 Published online Otober 8, 205 (http://www.sienepublishinggroup.om/j/ajsea) doi: 0.648/j.ajsea.2050406.2 ISSN: 2327-2473 (Print);

More information

Rapid, accurate particle tracking by calculation of radial symmetry centers

Rapid, accurate particle tracking by calculation of radial symmetry centers Rapid, aurate partile traing by alulation of radial symmetry enters Raghuveer Parthasarathy Supplementary Text and Figures Supplementary Figures Supplementary Figure 1 Supplementary Figure Supplementary

More information

Segmentation of brain MR image using fuzzy local Gaussian mixture model with bias field correction

Segmentation of brain MR image using fuzzy local Gaussian mixture model with bias field correction IOSR Journal of VLSI and Signal Proessing (IOSR-JVSP) Volume 2, Issue 2 (Mar. Apr. 2013), PP 35-41 e-issn: 2319 4200, p-issn No. : 2319 4197 Segmentation of brain MR image using fuzzy loal Gaussian mixture

More information

BioTechnology. An Indian Journal FULL PAPER. Trade Science Inc. Improvement of low illumination image enhancement algorithm based on physical mode

BioTechnology. An Indian Journal FULL PAPER. Trade Science Inc. Improvement of low illumination image enhancement algorithm based on physical mode [Type text] [Type text] [Type text] ISSN : 0974-7435 Volume 10 Issue 22 BioTehnology 2014 An Indian Journal FULL PAPER BTAIJ, 10(22), 2014 [13995-14001] Improvement of low illumination image enhanement

More information

Algorithms, Mechanisms and Procedures for the Computer-aided Project Generation System

Algorithms, Mechanisms and Procedures for the Computer-aided Project Generation System Algorithms, Mehanisms and Proedures for the Computer-aided Projet Generation System Anton O. Butko 1*, Aleksandr P. Briukhovetskii 2, Dmitry E. Grigoriev 2# and Konstantin S. Kalashnikov 3 1 Department

More information

Pipelined Multipliers for Reconfigurable Hardware

Pipelined Multipliers for Reconfigurable Hardware Pipelined Multipliers for Reonfigurable Hardware Mithell J. Myjak and José G. Delgado-Frias Shool of Eletrial Engineering and Computer Siene, Washington State University Pullman, WA 99164-2752 USA {mmyjak,

More information

Deep Rule-Based Classifier with Human-level Performance and Characteristics

Deep Rule-Based Classifier with Human-level Performance and Characteristics Deep Rule-Based Classifier with Human-level Performane and Charateristis Plamen P. Angelov 1,2 and Xiaowei Gu 1* 1 Shool of Computing and Communiations, Lanaster University, Lanaster, LA1 4WA, UK 2 Tehnial

More information

We P9 16 Eigenray Tracing in 3D Heterogeneous Media

We P9 16 Eigenray Tracing in 3D Heterogeneous Media We P9 Eigenray Traing in 3D Heterogeneous Media Z. Koren* (Emerson), I. Ravve (Emerson) Summary Conventional two-point ray traing in a general 3D heterogeneous medium is normally performed by a shooting

More information

Figure 1. LBP in the field of texture analysis operators.

Figure 1. LBP in the field of texture analysis operators. L MEHODOLOGY he loal inary pattern (L) texture analysis operator is defined as a gray-sale invariant texture measure, derived from a general definition of texture in a loal neighorhood. he urrent form

More information

Spatio-Temporal Naive-Bayes Nearest-Neighbor (ST-NBNN) for Skeleton-Based Action Recognition

Spatio-Temporal Naive-Bayes Nearest-Neighbor (ST-NBNN) for Skeleton-Based Action Recognition Spatio-Temporal Naive-Bayes Nearest-Neighbor () for Skeleton-Based Ation Reognition Junwu Weng Chaoqun Weng Junsong Yuan Shool of Eletrial and Eletroni Engineering Nanyang Tehnologial University, Singapore

More information

timestamp, if silhouette(x, y) 0 0 if silhouette(x, y) = 0, mhi(x, y) = and mhi(x, y) < timestamp - duration mhi(x, y), else

timestamp, if silhouette(x, y) 0 0 if silhouette(x, y) = 0, mhi(x, y) = and mhi(x, y) < timestamp - duration mhi(x, y), else 3rd International Conferene on Multimedia Tehnolog(ICMT 013) An Effiient Moving Target Traking Strateg Based on OpenCV and CAMShift Theor Dongu Li 1 Abstrat Image movement involved bakground movement and

More information

Learning Transferable Features with Deep Adaptation Networks

Learning Transferable Features with Deep Adaptation Networks Learning Transferable Features with Deep Adaptation Networks Mingsheng Long, Yue Cao, Jianmin Wang, Michael I. Jordan Presented by Changyou Chen October 30, 2015 1 Changyou Chen Learning Transferable Features

More information

Flow Demands Oriented Node Placement in Multi-Hop Wireless Networks

Flow Demands Oriented Node Placement in Multi-Hop Wireless Networks Flow Demands Oriented Node Plaement in Multi-Hop Wireless Networks Zimu Yuan Institute of Computing Tehnology, CAS, China {zimu.yuan}@gmail.om arxiv:153.8396v1 [s.ni] 29 Mar 215 Abstrat In multi-hop wireless

More information

Semi-Supervised Affinity Propagation with Instance-Level Constraints

Semi-Supervised Affinity Propagation with Instance-Level Constraints Semi-Supervised Affinity Propagation with Instane-Level Constraints Inmar E. Givoni, Brendan J. Frey Probabilisti and Statistial Inferene Group University of Toronto 10 King s College Road, Toronto, Ontario,

More information

The goal of machine learning

The goal of machine learning [appliations CORNER] Kilian Q. Weinberger, Fei Sha, and Lawrene K. Saul Convex Optimizations for Distane Metri Learning and Pattern Classifiation The goal of mahine learning is to build automated systems

More information

Disentangling Latent Space for VAE by Label Relevant/Irrelevant Dimensions

Disentangling Latent Space for VAE by Label Relevant/Irrelevant Dimensions Disentangling Latent Spae for VAE by Label Relevant/Irrelevant Dimensions Zhilin Zheng 1,2 Li Sun 1,2 1 Shanghai Key Laboratory of Multidimensional Information Proessing 2 East China Normal University

More information

Particle Swarm Optimization for the Design of High Diffraction Efficient Holographic Grating

Particle Swarm Optimization for the Design of High Diffraction Efficient Holographic Grating Original Artile Partile Swarm Optimization for the Design of High Diffration Effiient Holographi Grating A.K. Tripathy 1, S.K. Das, M. Sundaray 3 and S.K. Tripathy* 4 1, Department of Computer Siene, Berhampur

More information

Naïve Bayesian Rough Sets Under Fuzziness

Naïve Bayesian Rough Sets Under Fuzziness IJMSA: Vol. 6, No. 1-2, January-June 2012, pp. 19 25 Serials Publiations ISSN: 0973-6786 Naïve ayesian Rough Sets Under Fuzziness G. GANSAN 1,. KRISHNAVNI 2 T. HYMAVATHI 3 1,2,3 Department of Mathematis,

More information

A Fast Kernel-based Multilevel Algorithm for Graph Clustering

A Fast Kernel-based Multilevel Algorithm for Graph Clustering A Fast Kernel-based Multilevel Algorithm for Graph Clustering Inderjit Dhillon Dept. of Computer Sienes University of Texas at Austin Austin, TX 78712 inderjit@s.utexas.edu Yuqiang Guan Dept. of Computer

More information

Optimization of Two-Stage Cylindrical Gear Reducer with Adaptive Boundary Constraints

Optimization of Two-Stage Cylindrical Gear Reducer with Adaptive Boundary Constraints 5 JOURNAL OF SOFTWARE VOL. 8 NO. 8 AUGUST Optimization of Two-Stage Cylindrial Gear Reduer with Adaptive Boundary Constraints Xueyi Li College of Mehanial and Eletroni Engineering Shandong University of

More information

An Interactive-Voting Based Map Matching Algorithm

An Interactive-Voting Based Map Matching Algorithm Eleventh International Conferene on Mobile Data Management An Interative-Voting Based Map Mathing Algorithm Jing Yuan* University of Siene and Tehnology of China Hefei, China yuanjing@mail.ust.edu.n Yu

More information

HEXA: Compact Data Structures for Faster Packet Processing

HEXA: Compact Data Structures for Faster Packet Processing Washington University in St. Louis Washington University Open Sholarship All Computer Siene and Engineering Researh Computer Siene and Engineering Report Number: 27-26 27 HEXA: Compat Data Strutures for

More information

Diffusion Kernels on Graphs and Other Discrete Structures

Diffusion Kernels on Graphs and Other Discrete Structures Diffusion Kernels on Graphs and Other Disrete Strutures Risi Imre Kondor ohn Lafferty Shool of Computer Siene Carnegie Mellon University Pittsburgh P 523 US KONDOR@CMUEDU LFFERTY@CSCMUEDU bstrat The appliation

More information

Semantic Concept Detection Using Weighted Discretization Multiple Correspondence Analysis for Disaster Information Management

Semantic Concept Detection Using Weighted Discretization Multiple Correspondence Analysis for Disaster Information Management Semanti Conept Detetion Using Weighted Disretization Multiple Correspondene Analysis for Disaster Information Management Samira Pouyanfar and Shu-Ching Chen Shool of Computing and Information Sienes Florida

More information

Context-Aware Activity Modeling using Hierarchical Conditional Random Fields

Context-Aware Activity Modeling using Hierarchical Conditional Random Fields Context-Aware Ativity Modeling using Hierarhial Conditional Random Fields Yingying Zhu, Nandita M. Nayak, and Amit K. Roy-Chowdhury Abstrat In this paper, rather than modeling ativities in videos individually,

More information

Simulation of Crystallographic Texture and Anisotropie of Polycrystals during Metal Forming with Respect to Scaling Aspects

Simulation of Crystallographic Texture and Anisotropie of Polycrystals during Metal Forming with Respect to Scaling Aspects Raabe, Roters, Wang Simulation of Crystallographi Texture and Anisotropie of Polyrystals during Metal Forming with Respet to Saling Aspets D. Raabe, F. Roters, Y. Wang Max-Plank-Institut für Eisenforshung,

More information

A DYNAMIC ACCESS CONTROL WITH BINARY KEY-PAIR

A DYNAMIC ACCESS CONTROL WITH BINARY KEY-PAIR Malaysian Journal of Computer Siene, Vol 10 No 1, June 1997, pp 36-41 A DYNAMIC ACCESS CONTROL WITH BINARY KEY-PAIR Md Rafiqul Islam, Harihodin Selamat and Mohd Noor Md Sap Faulty of Computer Siene and

More information

Unsupervised Domain Adaptation by Domain Invariant Projection

Unsupervised Domain Adaptation by Domain Invariant Projection 2013 IEEE International Conference on Computer Vision Unsupervised Domain Adaptation by Domain Invariant Projection Mahsa Baktashmotlagh 1,3, Mehrtash T. Harandi 2,3, Brian C. Lovell 1, and Mathieu Salzmann

More information

Gradient based progressive probabilistic Hough transform

Gradient based progressive probabilistic Hough transform Gradient based progressive probabilisti Hough transform C.Galambos, J.Kittler and J.Matas Abstrat: The authors look at the benefits of exploiting gradient information to enhane the progressive probabilisti

More information

On - Line Path Delay Fault Testing of Omega MINs M. Bellos 1, E. Kalligeros 1, D. Nikolos 1,2 & H. T. Vergos 1,2

On - Line Path Delay Fault Testing of Omega MINs M. Bellos 1, E. Kalligeros 1, D. Nikolos 1,2 & H. T. Vergos 1,2 On - Line Path Delay Fault Testing of Omega MINs M. Bellos, E. Kalligeros, D. Nikolos,2 & H. T. Vergos,2 Dept. of Computer Engineering and Informatis 2 Computer Tehnology Institute University of Patras,

More information