Search For:

Displaying 1-50 out of 50 total
A fast mean-field method for large-scale high-dimensional data and its application in colonic polyp detection at CT colonography
Found in: Neural Networks, IEEE - INNS - ENNS International Joint Conference on
By Shijun Wang, Ronald M. Summers, Changshui Zhang
Issue Date:June 2009
pp. 3251-3258
In this paper, we propose a fast mean-field method called LHMF to handle probabilistic models of large-scale data in high dimensional space. By using diffusion map locally linear embedding method which is a non-linear dimensionality reduction method, we fi...
 
Beyond Banditron: A Conservative and Efficient Reduction for Online Multiclass Prediction with Bandit Setting Model
Found in: Data Mining, IEEE International Conference on
By Guangyun Chen, Gang Chen, Jianwen Zhang, Shuo Chen, Changshui Zhang
Issue Date:December 2009
pp. 71-80
In this paper, we consider a recently proposed supervised learning problem, called online multiclass prediction with bandit setting model. Aiming at learning from partial feedback of online classification results, i.e. “true” when the predicting label is r...
 
Efficient multi-label classification with hypergraph regularization
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Gang Chen, Jianwen Zhang, Fei Wang, Changshui Zhang, Yuli Gao
Issue Date:June 2009
pp. 1658-1665
Many computer vision applications, such as image classification and video indexing, are usually multi-label classification problems in which an instance can be assigned to more than one category. In this paper, we present a novel multi-label classification...
 
Nonlinear Dimensionality Reduction with Local Spline Embedding
Found in: IEEE Transactions on Knowledge and Data Engineering
By Shiming Xiang, Feiping Nie, Changshui Zhang, Chunxia Zhang
Issue Date:September 2009
pp. 1285-1298
This paper presents a new algorithm for Nonlinear Dimensionality Reduction (NLDR). Our algorithm is developed under the conceptual framework of compatible mapping. Each such mapping is a compound of a tangent space projection and a group of splines. Tangen...
 
A Unified Optimization Based Learning Method for Image Retrieval
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Hanghang Tong, Jingrui He, Mingjing Li, Wei-Ying Ma, Changshui Zhang, Hong-Jiang Zhang
Issue Date:June 2005
pp. 230-235
In this paper, an optimization based learning method is proposed for image retrieval from graph model point of view. Firstly, image retrieval is formulated as a regularized optimization problem, which simultaneously considers the constraints from low-level...
 
Learning No-Reference Quality Metric by Examples
Found in: Multi-Media Modeling Conference, International
By Hanghang Tong, Mingjing Li, Hong-Jiang Zhang, Changshui Zhang, Jingrui He, Wei-Ying Ma
Issue Date:January 2005
pp. 247-254
In this paper, a novel learning based method is proposed for No-Reference image quality assessment. Instead of examining the exact prior knowledge for the given type of distortion and finding a suitable way to represent it, our method aims to directly get ...
 
W-Boost and Its Application to Web Image Classification
Found in: Pattern Recognition, International Conference on
By Jingrui He, Mingjing Li, Hong-Jiang Zhang, Changshui Zhang
Issue Date:August 2004
pp. 148-151
When training data is not sufficient, boosting algorithms tend to overfit as more weak learners are combined to form a strong classifier. In this paper, we propose a new variant of RealBoost, called W-Boost, which is based on a novel weight update scheme a...
 
Blind Separation of Superimposed Moving Images Using Image Statistics
Found in: IEEE Transactions on Pattern Analysis and Machine Intelligence
By Kun Gai,Zhenwei Shi,Changshui Zhang
Issue Date:January 2012
pp. 19-32
We address the problem of blind separation of multiple source layers from their linear mixtures with unknown mixing coefficients and unknown layer motions. Such mixtures can occur when one takes photos through a transparent medium, like a window glass, and...
 
Homotopy Regularization for Boosting
Found in: Data Mining, IEEE International Conference on
By Zheng Wang, Yangqiu Song, Changshui Zhang
Issue Date:December 2010
pp. 1115-1120
In this paper, we present a homotopy regularization algorithm for boosting. We introduce a regularization term with adaptive weight into the boosting framework and compose a homotopy objective function. Optimization of this objective approximately composes...
 
Semi-Supervised Classification via Local Spline Regression
Found in: IEEE Transactions on Pattern Analysis and Machine Intelligence
By Shiming Xiang, Feiping Nie, Changshui Zhang
Issue Date:November 2010
pp. 2039-2053
This paper presents local spline regression for semi-supervised classification. The core idea in our approach is to introduce splines developed in Sobolev space to map the data points directly to be class labels. The spline is composed of polynomials and G...
 
Maximum Margin Clustering with Multivariate Loss Function
Found in: Data Mining, IEEE International Conference on
By Bin Zhao, James Kwok, Changshui Zhang
Issue Date:December 2009
pp. 637-646
This paper presents a simple but powerful extension of the maximum margin clustering (MMC) algorithm that optimizes multivariate performance measure specifically defined for clustering, including Normalized Mutual In- formation, Rand Index and F-measure. D...
 
Sparse Norm-Regularized Reconstructive Coefficients Learning
Found in: Data Mining, IEEE International Conference on
By Bin Liu, Shuo Chen, Mingjie Qian, Changshui Zhang
Issue Date:December 2009
pp. 854-859
Inspired by the fact that the final decision rule is mainly affected by a small subset of the training samples, i.e., Support Vector Machine(SVM) shows that the decision function relies on the few samples that are on or over the margin. We propose a new fr...
 
Unsupervised Maximum Margin Feature Selection with manifold regularization
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Bin Zhao, J. Kwok, Fei Wang, Changshui Zhang
Issue Date:June 2009
pp. 888-895
Feature selection plays a fundamental role in many pattern recognition problems. However, most efforts have been focused on the supervised scenario, while unsupervised feature selection remains as a rarely touched research topic. In this paper, we propose ...
 
Blind separation of superimposed images with unknown motions
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Kun Gai, Zhenwei Shi, Changshui Zhang
Issue Date:June 2009
pp. 1881-1888
We consider the blind separation of source layers from superimposed mixtures thereof, involving unknown motions and unknown mixing coefficients of layers in each mixture. Previous blind separation approaches for such problems assume motions to be uniform t...
 
Clustering with Local and Global Regularization
Found in: IEEE Transactions on Knowledge and Data Engineering
By Fei Wang, Changshui Zhang, Tao Li
Issue Date:December 2009
pp. 1665-1678
Clustering is an old research topic in data mining and machine learning. Most of the traditional clustering methods can be categorized as local or global ones. In this paper, a novel clustering method that can explore both the local and global information ...
 
Maximum Margin Embedding
Found in: Data Mining, IEEE International Conference on
By Bin Zhao, Fei Wang, Changshui Zhang
Issue Date:December 2008
pp. 1127-1132
We propose a new dimensionality reduction method called Maximum Margin Embedding (MME), which targets to projecting data samples into the most discriminative subspace, where clusters are most well-separated. Specifically, MME projects input patterns onto t...
 
Transductive object cutout
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Jingyu Cui, Qiong Yang, Fang Wen, Qiying Wu, Changshui Zhang, Luc Van Gool, Xiaoou Tang
Issue Date:June 2008
pp. 1-8
In this paper, we address the issue of transducing the object cutout model from an example image to novel image instances. We observe that although object and background are very likely to contain similar colors in natural images, it is much less probable ...
 
Blindly separating mixtures of multiple layers with spatial shifts
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Kun Gai, Zhenwei Shi, Changshui Zhang
Issue Date:June 2008
pp. 1-8
We address the problem of blindly separating mixtures of multiple layer images with unknown spatial shifts and mixing coefficients. Our proposed method can handle the over-determined, determined and under-determined cases where mixtures are more than, as m...
 
Normalized tree partitioning for image segmentation
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Jingdong Wang, Yangqing Jia, Xian-Sheng Hua, Changshui Zhang, Long Quan
Issue Date:June 2008
pp. 1-8
In this paper, we propose a novel graph based clustering approach with satisfactory clustering performance and low computational cost. It consists of two main steps: tree fitting and partitioning. We first introduce a probabilistic method to fit a tree to ...
 
Collaborative Filtering Using Orthogonal Nonnegative Matrix Tri-factorization
Found in: Data Mining Workshops, International Conference on
By Gang Chen, Fei Wang, Changshui Zhang
Issue Date:October 2007
pp. 303-308
Collaborative filtering aims at predicting a test user's ratings for new items by integrating other like-minded users' rating information. Traditional collaborative filter- ing methods usually suffer from two fundamental problems: sparsity and scalability....
 
Multilevel Belief Propagation for Fast Inference on Markov Random Fields
Found in: Data Mining, IEEE International Conference on
By Liang Xiong, Fei Wang, Changshui Zhang
Issue Date:October 2007
pp. 371-380
Graph-based inference plays an important role in many mining and learning tasks. Among all the solvers for this problem, belief propagation (BP) provides a general and efficient way to derive approximate solutions. However, for large scale graphs the compu...
 
Simultaneous Heterogeneous Data Clustering Based on Higher Order Relationships
Found in: Data Mining Workshops, International Conference on
By Shouchun Chen, Fei Wang, Changshui Zhang
Issue Date:October 2007
pp. 387-392
Co-clustering on heterogeneous data has attracted more and more attention in web mining and information retrieval. The clustering approaches for two type heterogeneous data (bi-type co-clustering) have been well studied in the lit- erature. However, the wo...
 
Label Propagation through Linear Neighborhoods
Found in: IEEE Transactions on Knowledge and Data Engineering
By Fei Wang, Changshui Zhang
Issue Date:January 2008
pp. 55-67
In many practical data mining applications such as text classification, unlabeled training examples are readily available but labeled ones are fairly expensive to obtain. Therefore, semi-supervised learning algorithms have aroused considerable interests fr...
 
Feature Extraction by Maximizing the Average Neighborhood Margin
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Fei Wang, Changshui Zhang
Issue Date:June 2007
pp. 1-8
A novel algorithm called Average Neighborhood Margin Maximization (ANMM) is proposed for supervised linear feature extraction. For each data point, ANMM aims at pulling the neighboring points with the same class label to-wards it as near as possible, while...
 
Optimal Dimensionality Discriminant Analysis and Its Application to Image Recognition
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Feiping Nie, Shiming Xiang, Yangqiu Song, Changshui Zhang
Issue Date:June 2007
pp. 1-8
Dimensionality reduction is an important issue when facing high-dimensional data. For supervised dimensionality reduction, Linear Discriminant Analysis (LDA) is one of the most popular methods and has been successfully applied in many classification proble...
 
Discriminant Additive Tangent Spaces for Object Recognition
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Liang Xiong, Jianguo Li, Changshui Zhang
Issue Date:June 2007
pp. 1-8
Pattern variation is a major factor that affects the performance of recognition systems. In this paper, a novel manifold tangent modeling method called Discriminant Additive Tangent Spaces (DATS) is proposed for invariant pattern recognition. In DATS, intr...
 
Semi-Supervised Classification Using Linear Neighborhood Propagation
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Fei Wang, Changshui Zhang, Helen C. Shen, Jingdong Wang
Issue Date:June 2006
pp. 160-167
In this paper, we address the general problem of learning from both labeled and unlabeled data. Based on the reasonable assumption that the label of each data can be linearly reconstructed from its neighbors? labels, we develop a novel approach, called Lin...
 
A Discriminative Method For Semi-Automated Tumorous Tissues Segmentation of MR Brain Images
Found in: Computer Vision and Pattern Recognition Workshop
By Yangqiu Song, Changshui Zhang, Jianguo Lee, Fei Wang
Issue Date:June 2006
pp. 79
This paper introduces a discriminative method for semiautomated segmentation of the tumorous tissues. Due to the large data of 3D MR brain images and the blurry boundary of the pathological tissues, the segmentation is difficult. A non-parametric Bayesian ...
 
Relevant Linear Feature Extraction Using Side-information and Unlabeled Data
Found in: Pattern Recognition, International Conference on
By Fei Wu, Yonglei Zhou, Changshui Zhang
Issue Date:August 2004
pp. 582-585
No summary available.
 
Classifier Combination based on Active Learning
Found in: Pattern Recognition, International Conference on
By Xing Yi, Zhongbao Kou, Changshui Zhang
Issue Date:August 2004
pp. 184-187
In this paper, we propose Classifier Combination based on Active Learning, which deals with the design of classifier combination systems as training a combiner at the aggregation level and introduces SVM active learning into the design of this multi-catego...
 
Active Morphable Model: An Efficient Method for Face Analysis
Found in: Automatic Face and Gesture Recognition, IEEE International Conference on
By Xun Xu, Changshui Zhang, Thomas S. Huang
Issue Date:May 2004
pp. 837
Multidimensional Morphable Model is a powerful model to analyze and synthesize human faces. However, the stochastic gradient descent algorithm adopted to match the Morphable Model to a novel face image is not efficient enough. In this paper, a very efficie...
 
A Fully Automated Face Recognition System Under Different Conditions
Found in: Pattern Recognition, International Conference on
By Hui Peng, Changshui Zhang, Zhaoqi Bian
Issue Date:August 1998
pp. 1223
No summary available.
 
Evolutionary hierarchical dirichlet processes for multiple correlated time-varying corpora
Found in: Proceedings of the 16th ACM SIGKDD international conference on Knowledge discovery and data mining (KDD '10)
By Changshui Zhang, Jianwen Zhang, Shixia Liu, Yangqiu Song
Issue Date:July 2010
pp. 1079-1088
Mining cluster evolution from multiple correlated time-varying text corpora is important in exploratory text analytics. In this paper, we propose an approach called evolutionary hierarchical Dirichlet processes (EvoHDP) to discover interesting cluster evol...
     
Manifold-ranking based image retrieval
Found in: Proceedings of the 12th annual ACM international conference on Multimedia (MULTIMEDIA '04)
By Changshui Zhang, Hanghang Tong, Hong-Jiang Zhang, Jingrui He, Mingjing Li
Issue Date:October 2004
pp. 9-16
In this paper, we propose a novel transductive learning framework named manifold-ranking based image retrieval (MRBIR). Given a query image, MRBIR first makes use of a manifold ranking algorithm to explore the relationship among all the data points in the ...
     
Mean version space: a new active learning method for content-based image retrieval
Found in: Proceedings of the 6th ACM SIGMM international workshop on Multimedia information retrieval (MIR '04)
By Changshui Zhang, Hanghang Tong, Hong-Jiang Zhang, Jingrui He, Mingjing Li
Issue Date:October 2004
pp. 15-22
In content-based image retrieval, relevance feedback has been introduced to narrow the gap between low-level image feature and high-level semantic concept. Furthermore, to speed up the convergence to the query concept, several active learning methods have ...
     
Robust multi-task feature learning
Found in: Proceedings of the 18th ACM SIGKDD international conference on Knowledge discovery and data mining (KDD '12)
By Changshui Zhang, Jieping Ye, Pinghua Gong
Issue Date:August 2012
pp. 895-903
Multi-task learning (MTL) aims to improve the performance of multiple related tasks by exploiting the intrinsic relationships among them. Recently, multi-task feature learning algorithms have received increasing attention and they have been successfully ap...
     
Mining interlacing manifolds in high dimensional spaces
Found in: Proceedings of the 2011 ACM Symposium on Applied Computing (SAC '11)
By Changshui Zhang, Shigeo Abe, Takeshi Takahashi, Tao Ban, Youki Kadobayashi
Issue Date:March 2011
pp. 942-949
Real world data are often composed of conceptually meaningful subspaces, e.g., for portraits in a facial image database, the illumination factor corresponds to a nonlinear subspace and the rotation factor corresponds to another. The interlacement of these ...
     
Learning to rank tags
Found in: Proceedings of the ACM International Conference on Image and Video Retrieval (CIVR '10)
By Changshui Zhang, Jiashi Feng, Shuicheng Yan, Zheng Wang
Issue Date:July 2010
pp. 42-49
Social images sharing websites, such as Flickr and Picasa, are becoming very popular nowadays. Users are generally recommended to annotate images with free tags, yet these tags are orderless, and thus quite limited for applications like image search, retri...
     
Efficient multi-class unlabeled constrained semi-supervised SVM
Found in: Proceeding of the 18th ACM conference on Information and knowledge management (CIKM '09)
By Changshui Zhang, Feiping Nie, Mingjie Qian
Issue Date:November 2009
pp. 1665-1668
Semi-supervised learning has been successfully applied to many fields such as knowledge management, information retrieval and data mining as it can utilize both labeled and unlabeled data. In this paper, we propose a general semi-supervised framework for m...
     
Instance- and bag-level manifold regularization for aggregate outputs classification
Found in: Proceeding of the 18th ACM conference on Information and knowledge management (CIKM '09)
By Bin Liu, Changshui Zhang, Mingjie Qian, Shuo Chen
Issue Date:November 2009
pp. 1593-1596
Aggregate outputs learning differs from the classical supervised learning setting in that, training samples are packed into bags with only the aggregate outputs (labels for classification or real values for regression) known. This setting of the problem is...
     
Finding image exemplars using fast sparse affinity propagation
Found in: Proceeding of the 16th ACM international conference on Multimedia (MM '08)
By Changshui Zhang, Jingdong Wang, Xian-Sheng Hua, Yangqing Jia
Issue Date:October 2008
pp. 40-42
In this paper, we propose a novel approach to organize image search results obtained from state-of-the-art image search engines in order to improve user experience. We aim to discover exemplars from search results and simultaneously group the images. The e...
     
Semi-supervised metric learning by maximizing constraint margin
Found in: Proceeding of the 17th ACM conference on Information and knowledge mining (CIKM '08)
By Changshui Zhang, Fei Wang, Shouchun Chen, Tao Li
Issue Date:October 2008
pp. 1001-1001
Distance metric learning is an old problem that has been researched in the supervised learning field for a very long time. In this paper, we consider the problem of learning a proper distance metric under the guidance of some weak supervisory information. ...
     
Semi-supervised ranking aggregation
Found in: Proceeding of the 17th ACM conference on Information and knowledge mining (CIKM '08)
By Changshui Zhang, Fei Wang, Shouchun Chen, Yaangqiu Song
Issue Date:October 2008
pp. 1001-1001
Ranking aggregation is important in data mining and information retrieval. In this paper, we proposed a semi-supervised ranking aggregation method, in which the order of several item pairs are labeled as side information. The core idea is to learn a rankin...
     
Cuts3vm: a fast semi-supervised svm algorithm
Found in: Proceeding of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining (KDD '08)
By Bin Zhao, Changshui Zhang, Fei Wang
Issue Date:August 2008
pp. 5-6
Semi-supervised support vector machine (S3VM) attempts to learn a decision boundary that traverses through low data density regions by maximizing the margin over labeled and unlabeled examples. Traditionally, S3VM is formulated as a non-convex integer prog...
     
Efficient multiclass maximum margin clustering
Found in: Proceedings of the 25th international conference on Machine learning (ICML '08)
By Bin Zhao, Changshui Zhang, Fei Wang
Issue Date:July 2008
pp. 1248-1255
This paper presents a cutting plane algorithm for multiclass maximum margin clustering (MMC). The proposed algorithm constructs a nested sequence of successively tighter relaxations of the original MMC problem, and each optimization problem in this sequenc...
     
Regularized clustering for documents
Found in: Proceedings of the 30th annual international ACM SIGIR conference on Research and development in information retrieval (SIGIR '07)
By Changshui Zhang, Fei Wang, Tao Li
Issue Date:July 2007
pp. 95-102
In recent years, document clustering has been receiving more and more attentions as an important and fundamental technique for unsupervised document organization, automatictopic extraction, and fast information retrieval or filtering. In this paper, we pro...
     
Label propagation through linear neighborhoods
Found in: Proceedings of the 23rd international conference on Machine learning (ICML '06)
By Changshui Zhang, Fei Wang
Issue Date:June 2006
pp. 985-992
A novel semi-supervised learning approach is proposed based on a linear neighborhood model, which assumes that each data point can be linearly reconstructed from its neighborhood. Our algorithm, named Linear Neighborhood Propagation (LNP), can propagate th...
     
Multiple random walk and its application in content-based image retrieval
Found in: Proceedings of the 7th ACM SIGMM international workshop on Multimedia information retrieval (MIR '05)
By Changshui Zhang, Hanghang Tong, Jingrui He, Mingjing Li, Wei-Ying Ma
Issue Date:November 2005
pp. 151-158
In this paper, we propose a transductive learning method for content-based image retrieval: Multiple Random Walk (MRW). Its basic idea is to construct two generative models by means of Markov random walks, one for images relevant to the query concept and t...
     
Graph based multi-modality learning
Found in: Proceedings of the 13th annual ACM international conference on Multimedia (MULTIMEDIA '05)
By Changshui Zhang, Hanghang Tong, Jingrui He, Mingjing Li, Wei-Ying Ma
Issue Date:November 2005
pp. 862-871
To better understand the content of multimedia, a lot of research efforts have been made on how to learn from multi-modal feature. In this paper, it is studied from a graph point of view: each kind of feature from one modality is represented as one indepen...
     
Probabilistic tangent subspace: a unified view
Found in: Twenty-first international conference on Machine learning (ICML '04)
By Changshui Zhang, Jianguo Lee, Jingdong Wang, Zhaoqi Bian
Issue Date:July 2004
pp. 182-182
Tangent Distance (TD) is one classical method for invariant pattern classification. However, conventional TD need pre-obtain tangent vectors, which is difficult except for image objects. This paper extends TD to more general pattern classification tasks. T...
     
 1