[Show abstract][Hide abstract] ABSTRACT: Model-based approaches and in particular finite mixture models are widely used for data clustering which is a crucial step in several applications of practical importance. Indeed, many pattern recognition, computer vision and image processing applications can be approached as feature space clustering problems. For complex high-dimensional data, however, the use of these approaches presents several challenges such as the presence of many irrelevant features which may affect the speed and also compromise the accuracy of the used learning algorithm. Another problem is the presence of outliers which potentially influence the resulting model’s parameters. For this purpose, we propose and discuss an algorithm that partitions a given data set without a priori information about the number of clusters, the saliency of the features or the number of outliers. We illustrate the performance of our approach using different applications involving synthetic data, real data and objects shape clustering.
Expert Systems with Applications 05/2012; 39(7):6641–6656. · 1.85 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: In this paper, we formalize content-based image suggestion (CBIS) as a Bayesian prediction problem. In CBIS, users provide the rating of images according to both their long-term needs and the contextual situation, such as time and place, to which they belong. Therefore, a CBIS model is defined to fit the distribution of the data in order to predict relevant images for a given user. Generally, CBIS becomes challenging when only a small amount of data is available such as in the case of “new users” and “new images.” The Bayesian predictive approach is an effective solution to such a problem. In addition, this approach offers efficient means to select highly rated and diversified suggestions in conformance with theories in consumer psychology. Experiments on a real data set show the merits of our approach in terms of image suggestion accuracy and efficiency.
IEEE transactions on neural networks and learning systems 01/2012; 23(8):1242-1253. · 3.77 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: A method for simultaneous non-Gaussian data clustering, feature selection and outliers rejection is proposed in this paper.
The proposed approach is based on finite generalized Dirichlet mixture models learned within a framework including expectation-maximization
updates for model parameters estimation and minimum message length criterion for model selection. Through a challenging application
involving texture images discrimination, it is demonstrated that the developed procedure performs effectively in avoiding
outliers and selecting relevant features.
Pattern Recognition and Machine Intelligence - 4th International Conference, PReMI 2011, Moscow, Russia, June 27 - July 1, 2011. Proceedings; 01/2011
[Show abstract][Hide abstract] ABSTRACT: In this letter, we propose a clustering model that efficiently mitigates image and video under/over-segmentation by combining generalized Gaussian mixture modeling and feature selection. The model has flexibility to accurately represent heavy-tailed image/video histograms, while automatically discarding uninformative features, leading to better discrimination and localization of regions in high-dimensional spaces. Experimental results on a database of real-world images and videos showed us the effectiveness of the proposed approach.
IEEE Transactions on Circuits and Systems for Video Technology 11/2010; · 1.82 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: In this paper we investigate the integration of feature selection in segmentation through an unsupervised learning approach. We propose a clustering algorithm that efficiently mitigates image under/over-segmentation, by combining generalized Gaussian mixture modeling and feature selection. The algorithm is based on generalized Gaussian mixture modeling which is less prone to region number over-estimation in case of noisy and heavy-tailed image distributions. On the other hand, our feature selection mechanism allows to automatically discard uninformative features, which leads to better discrimination and localization of regions in high-dimensional spaces. Experimental results on a large database of real-world images showed us the effectiveness of the proposed approach.
Computer and Robot Vision (CRV), 2010 Canadian Conference on; 07/2010
[Show abstract][Hide abstract] ABSTRACT: Content-based image suggestion (CBIS) addresses the satisfaction of users long-term needs for “relevant” and “novel” images. In this paper, we present VCC-FMM, a flexible mixture model that clusters both images and users into separate groups. Then, we propose long-term relevance feedback to maintain accurate modeling of growing image collections and changing user long-term needs over time. Experiments on a real data set show merits of our approach in terms of image suggestion accuracy and efficiency.
[Show abstract][Hide abstract] ABSTRACT: This paper presents a new generalized Dirichlet (GD) mixture model to address the challenging problem of clustering multidimensional data sets on different feature subsets. We approximate class-conditional distributions of mixture components to define binary relevance of features at the level of clusters. We consider a relevant feature as the one providing the knowledge to assign data points in the cluster. Then, we define a new message length objective to learn the model and select both feature subsets and the number of components. The proposed method is general comparatively with existing feature selection and subspace clustering models. In addition, it selects for each cluster only relevant and statistically independent features in a linear time of the number of observations and dimensions. Experiments on synthetic data and in unsupervised image categorization show the merits of our approach.
[Show abstract][Hide abstract] ABSTRACT: This paper presents an unsupervised approach for feature selection and extraction in mixtures of generalized Dirichlet (GD) distributions. Our method defines a new mixture model that is able to extract independent and non-Gaussian features without loss of accuracy. The proposed model is learned using the Expectation-Maximization algorithm by minimizing the message length of the data set. Experimental results show the merits of the proposed methodology in the categorization of object images.
IEEE Transactions on Pattern Analysis and Machine Intelligence 09/2009; 31(8):1429-43. · 4.80 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: In this paper, we propose a probabilistic framework for efficient retrieval and indexing of image collections. This framework uncovers the hierarchical structure underlying the collection from image features based on a hybrid model that combines both generative and discriminative learning. We adopt the generalized Dirichlet mixture and maximum likelihood for the generative learning in order to estimate accurately the statistical model of the data. Then, the resulting model is refined by a new discriminative likelihood that enhances the power of relevant features. Consequently, this new model is suitable for modeling high-dimensional data described by both semantic and low-level (visual) features. The semantic features are defined according to a known ontology while visual features represent the visual appearance such as color, shape, and texture. For validation purposes, we propose a new visual feature which has nice invariance properties to image transformations. Experiments on the Microsoft's collection (MSRCID) show clearly the merits of our approach in both retrieval and indexing.
[Show abstract][Hide abstract] ABSTRACT: This paper presents a Bayesian approach to address two important issues of image recommendation that are: (1) change in long-term
needs of users and (2) evolution of image collections. Users are offered a new interaction modality which allows them to provide
either positive or negative relevance feedback (RF) data to express their recent needs. Then, an efficient variational Online
learning algorithm updates both user and product collection models by favoring recent RF data. The proposed method is general
and can be applied in collaborative filtering. Experimental results demonstrate the importance of maintaining most up-to-date
user models on the rating’s prediction accuracy.
Advances in Knowledge Discovery and Data Mining, 13th Pacific-Asia Conference, PAKDD 2009, Bangkok, Thailand, April 27-30, 2009, Proceedings; 01/2009
[Show abstract][Hide abstract] ABSTRACT: Existing recommender systems provide an elegant solution to the information overload in current digital libraries such as the Internet archive. Nowadays, the sensors that capture the user's contextual information such as the location and time are become available and have raised a need to personalize recommendations for each user according to his/her changing needs in different contexts. In addition, visual documents have richer textual and visual information that was not exploited by existing recommender systems. In this paper, we propose a new framework for context-aware recommendation of visual documents by modeling the user needs, the context and also the visual document collection together in a unified model. We address also the user's need for diversified recommendations. Our pilot study showed the merits of our approach in content based image retrieval.
IEEE Transactions on Multimedia 02/2008; · 1.75 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: We present in this paper a new approach for unsupervised feature selection for non Gaussian data controlled by a finite mixture of generalized Dirichlet distributions. We model each feature by a mixture of two Beta distributions: one relevant and depends on component labels while the second is uninformative for the clustering. The relevance of each feature is then quantified by the mixture weight associated to the relevant Beta distribution. Experiments in summarizing image collections have shown the effectiveness of our approach.
Machine Learning for Signal Processing, 2007 IEEE Workshop on; 09/2007
[Show abstract][Hide abstract] ABSTRACT: Content based image retrieval systems provide techniques for representing, indexing and searching images. They address only
the user’s short term needs expressed as queries. From the importance of the visual information in many applications such
as advertisements and security, we motivate in this paper, the Content Based Image Suggestion. It targets the user’s long term needs as a recommendation of products based on the user preferences in different situations,
and on the visual content of images. We propose a generative model in which the visual features and users are clustered into
separate classes. We identify the number of both user and image classes with the simultaneous selection of relevant visual
features. The goal is to ensure an accurate prediction of ratings for multidimensional images. This model is learned using
the minimum message length approach. Experiments with an image collection showed the merits of our approach.
Knowledge Discovery in Databases: PKDD 2007, 11th European Conference on Principles and Practice of Knowledge Discovery in Databases, Warsaw, Poland, September 17-21, 2007, Proceedings; 01/2007
[Show abstract][Hide abstract] ABSTRACT: Content-based image suggestion (CBIS) targets the recommendation of products based on user preferences on the visual content of images. In this paper, we mo- tivate both feature selection and model order identification as two key issues for a successful CBIS. We propose a generative model in which the visual features and users are clustered into separate classes. We identify the number of both user and image classes with the simultaneous selection of relevant visual features us- ing the message length approach. The goal is to ensure an accurate prediction of ratings for multidimensional non-Gaussian and continuous image descriptors. Experiments on a collected data have demonstrated the merits of our approach.
Advances in Neural Information Processing Systems 20, Proceedings of the Twenty-First Annual Conference on Neural Information Processing Systems, Vancouver, British Columbia, Canada, December 3-6, 2007; 01/2007
[Show abstract][Hide abstract] ABSTRACT: We present in this paper a new approach for unsupervised feature selection for non Gaussian data controlled by a finite mixture
of generalized Dirichlet distributions. We model each feature by a mixture of two Beta distributions: one relevant and depends
on component labels while the second distribution is uninformative for the clustering. The relevance of each feature is then
quantified by the mixture weight associated to the relevant Beta distribution. Experiments in summarizing image collections
have shown the merits of our approach.
Image Analysis and Recognition, 4th International Conference, ICIAR 2007, Montreal, Canada, August 22-24, 2007, Proceedings; 01/2007
[Show abstract][Hide abstract] ABSTRACT: In this paper we present a new approach for interacting with visual document collections. We propose to model user preferences
related to visual documents in order to recommend relevant content according to the user’s profile. We have formulated problem
as prediction problem and we propose VC-Aspect our flexible mixture model which handles implicit associations between users
and the visual features of images. We have implemented the model within a CBIR system and results showed that such approach
reduced greatly the page zero problem especially for small devices such as smart-phones and PDAs.
[Show abstract][Hide abstract] ABSTRACT: In this paper, we propose a novel generative graphical model for collaborative filtering of visual content. The preferences
of the ”like-minded” users are modelled in order to predict the relevance of visual documents represented by their visual
features. We formulate the problem using a probabilistic latent variable model where user’s preferences and items’ classes
are combined into a unified framework in order to provide an accurate and a generative model that overcomes the new item problem,
generally encountered in traditional collaborative filtering systems.
Advances in Data Mining, Applications in Medicine, Web Mining, Marketing, Image and Signal Mining, 6th Industrial Conference on Data Mining, ICDM 2006, Leipzig, Germany, July 14-15, 2006, Proceedings; 01/2006
[Show abstract][Hide abstract] ABSTRACT: This paper presents a generative graphical model (VC-Aspect) for filtering visual documents such as images. The proposed VC-Aspect
extends the well-known Aspect model and combines both content based and collaborative filtering approaches in a unified framework.
Instead of considering item indices in the model such as model-based collaborative filtering techniques, we use visual features
in describing visual documents. This allows the model to predict ratings for new visual documents with the same set of parameters.
Experimental results show the usefulness of such an approach in a real life application such as the content based image retrieval.
Image Analysis and Recognition, Third International Conference, ICIAR 2006, Póvoa de Varzim, Portugal, September 18-20, 2006, Proceedings, Part I; 01/2006
[Show abstract][Hide abstract] ABSTRACT: Collaborative filtering (CF) has proven its effi- ciency in recommending general purpose items such as movies, news, scientific papers etc. It has been noticed in literature that CF techniques pre- dict poorly ratings with sparse data and are unable to recommend non rated items. Other CF tech- niques which use the content of items have proven their efficiency in filtering text documents. This paper presents a new model-based CF approach for visual document recommendation which uses the low level features of visual documents in or- der to overcome the data sparsity and new item problems. Also, a new recommendation tech- nique which considers the diversity and novelty of recommended items for a given user is pre- sented. In-lab experiments showed the accuracy of the rating's prediction and the usefulness of the proposed approach in content based image re- trieval.