• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.7 no.3
• /
• pp.522-538
• /
• 2013

Content based image retrieval has become an increasingly important research topic for its wide application. It is highly challenging when facing to large-scale database with large variance. The retrieval systems rely on a key component, the predefined or learned similarity measures over images. We note that, the similarity measures can be potential improved if the data distribution information is exploited using a more sophisticated way. In this paper, we propose a similarity measure learning approach for image retrieval. The similarity measure, so called Fisher kernel, is derived from the probabilistic distribution of images and is the function over observed data, hidden variable and model parameters, where the hidden variables encode high level information which are powerful in discrimination and are failed to be exploited in previous methods. We further propose a discriminative learning method for the similarity measure, i.e., encouraging the learned similarity to take a large value for a pair of images with the same label and to take a small value for a pair of images with distinct labels. The learned similarity measure, fully exploiting the data distribution, is well adapted to dataset and would improve the retrieval system. We evaluate the proposed method on Corel-1000, Corel5k, Caltech101 and MIRFlickr 25,000 databases. The results show the competitive performance of the proposed method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.7 no.5
• /
• pp.1252-1271
• /
• 2013

Automatic image annotation has become an increasingly important research topic owing to its key role in image retrieval. Simultaneously, it is highly challenging when facing to large-scale dataset with large variance. Practical approaches generally rely on similarity measures defined over images and multi-label prediction methods. More specifically, those approaches usually 1) leverage similarity measures predefined or learned by optimizing for ranking or annotation, which might be not adaptive enough to datasets; and 2) predict labels separately without taking the correlation of labels into account. In this paper, we propose a method for image annotation through collaborative similarity metric learning from dataset and modeling the label correlation of the dataset. The similarity metric is learned by simultaneously optimizing the 1) image ranking using structural SVM (SSVM), and 2) image annotation using correlated label propagation, with respect to the similarity metric. The learned similarity metric, fully exploiting the available information of datasets, would improve the two collaborative components, ranking and annotation, and sequentially the retrieval system itself. We evaluated the proposed method on Corel5k, Corel30k and EspGame databases. The results for annotation and retrieval show the competitive performance of the proposed method.

• Journal of KIISE:Software and Applications
• /
• v.31 no.11
• /
• pp.1519-1533
• /
• 2004

When Web-based special retrieval systems for scientific field extremely restrict the expression of user's information request, the process of the information content analysis and that of the information acquisition become inconsistent. In this paper, we apply the fuzzy retrieval model to solve the high time complexity of the retrieval system by constructing a reduced term set for the term's relatively importance degree. Furthermore, we perform a cluster retrieval to reflect the user's Query exactly through the similarity relation matrix satisfying the characteristics of the fuzzy compatibility relation. We have proven the performance of a proposed re-ranking model based on the similarity union of the fuzzy retrieval model and the document cluster retrieval model.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.12 no.4
• /
• pp.324-328
• /
• 2012

With the recent advances in information retrieval, cross-media retrieval has been attracting lot of attention, but several issues remain problems such as constructing effective correlations, calculating similarity between different kinds of media objects. To gain better cross-media retrieval performance, it is crucial to mine the semantic correlations among the heterogeneous multimedia data. This paper introduces a new method for cross-media retrieval which uses ontology to organize different media objects. The experiment results show that the proposed method is effective in cross-media retrieval.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.8
• /
• pp.837-841
• /
• 2010

This paper proposes an effective similarity measure for content-based image retrieval using MPEG-7 DCD. The proposed method can measure the similarity of images with the percentage of dominant colors extracted from images. As the result of experiments, we achieved a significant improvement of 18.92% with global DCD and 47.22% with local DCD in ANMRR than the result by QHDM. This result shows that the proposed method is an effective similarity measure for content-based image retrieval. Especially, our method is useful for region-based image retrieval.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 1997.06a
• /
• pp.137-142
• /
• 1997

One of the inherent problems in constructing the sign language dictionary is how to make retrieval and comparison operations on the visual database of signs. This paper describes a retrieval method, especifically for Japanese signs. This method has a useful capability for flexible retrieval of the sign from a bilingual dictionary. Our method can retrieve similar signs to the given input. The retrieval mechanism is essentially based on similarity between the given verbal description and verbal descriptions in a retrieval database. The similarity measure of verbal descriptions can be considered as the approximations for the similarity of sign motion images. As a results of our experiment, the success ratio of the retrievals is 96% in averages.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.4
• /
• pp.1424-1440
• /
• 2015

It is a challenging problem to search the intended images from a large number of candidates. Content based image retrieval (CBIR) is the most promising way to tackle this problem, where the most important topic is to measure the similarity of images so as to cover the variance of shape, color, pose, illumination etc. While previous works made significant progresses, their adaption ability to dataset is not fully explored. In this paper, we propose a similarity learning method on the basis of probabilistic generative model, i.e., probabilistic latent semantic analysis (PLSA). It first derives Fisher kernel, a function over the parameters and variables, based on PLSA. Then, the parameters are determined through simultaneously maximizing the log likelihood function of PLSA and the retrieval performance over the training dataset. The main advantages of this work are twofold: (1) deriving similarity measure based on PLSA which fully exploits the data distribution and Bayes inference; (2) learning model parameters by maximizing the fitting of model to data and the retrieval performance simultaneously. The proposed method (PLSA-FK) is empirically evaluated over three datasets, and the results exhibit promising performance.

• Journal of Korea Multimedia Society
• /
• v.7 no.5
• /
• pp.617-632
• /
• 2004

Video Data includes moving objects which change spatial positions as time goes by. In this paper, we propose a new modeling method for a moving object contained in the video data. In order to effectively retrieve moving objects, the proposed modeling method represents the spatial position and the size of a moving object. It also represents the visual features and the trajectory by considering direction, distance and speed or moving objects as time goes by. Therefore, It allows various types of retrieval such as visual feature based similarity retrieval, distance based similarity retrieval and trajectory based similarity retrieval and their mixed type of weighted retrieval.

• Journal of Information Processing Systems
• /
• v.14 no.1
• /
• pp.32-55
• /
• 2018

The digitization of music has seen a considerable increase in audience size from a few localized listeners to a wider range of global listeners. At the same time, the digitization brings the challenge of smoothly retrieving music from large databases. To deal with this challenge, many systems which support the smooth retrieval of musical data have been developed. At the computational level, a query music piece is compared with the rest of the music pieces in the database. These systems, music information retrieval (MIR systems), work for various applications such as general music retrieval, plagiarism detection, music recommendation, and musicology. This paper mainly addresses two parts of the MIR research area. First, it presents a general overview of MIR, which will examine the history of MIR, the functionality of MIR, application areas of MIR, and the components of MIR. Second, we will investigate music similarity measurement methods, where we provide a comparative analysis of state of the art methods. The scope of this paper focuses on comparative analysis of the accuracy and efficiency of a few key MIR systems. These analyses help in understanding the current and future challenges associated with the field of MIR systems and music similarity measures.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.6
• /
• pp.2527-2545
• /
• 2016

Measuring the similarity of given samples is a key problem of recognition, clustering, retrieval and related applications. A number of works, e.g. kernel method and metric learning, have been contributed to this problem. The challenge of similarity learning is to find a similarity robust to intra-class variance and simultaneously selective to inter-class characteristic. We observed that, the similarity measure can be improved if the data distribution and hidden semantic information are exploited in a more sophisticated way. In this paper, we propose a similarity learning approach for retrieval and recognition. The approach, termed as LDA-FEK, derives free energy kernel (FEK) from Latent Dirichlet Allocation (LDA). First, it trains LDA and constructs kernel using the parameters and variables of the trained model. Then, the unknown kernel parameters are learned by a discriminative learning approach. The main contributions of the proposed method are twofold: (1) the method is computationally efficient and scalable since the parameters in kernel are determined in a staged way; (2) the method exploits data distribution and semantic level hidden information by means of LDA. To evaluate the performance of LDA-FEK, we apply it for image retrieval over two data sets and for text categorization on four popular data sets. The results show the competitive performance of our method.