Abstract not found

Abstract not found

Singular value decomposition (SVD) is a general-purpose mathematical analysis tool that has been used in a variety of information-retrieval applications. As the size and complexity of retrieval collections increase, it is crucial for our analysis tools to scale accordingly. To this end, we have studied the application of a new theoretically justified SVD approximation algorithm to the problem of text retrieval. We show that, in the case of latent semantic indexing, we can achieve near optimal approximations of the exact SVD using considerably less computation by using an appropriate distribution to sample the documents we include in our SVD analysis.

Image Retrieval (IR) is one of the most exciting and fastestgrowing research areas in the field of multimedia technology. We present here a highlight of recent research for IR. Some trends and probable future research directions are presented. We expose the major problems that we have recognized: the lack of a good measurement of visual similarity, the little importance accorded to user interaction and feedback, and the neglect of spatial information. Answering these concerns, we describe the solutions implemented by recent IR systems. We also present the current image retrieval projects in our laboratory, which are motivated to a large extent by these same considerations. 1

Histograms are commonly used in content-based image retrieval systems to represent the distributions of colors in images. It is a common understanding that histograms that adapt to images can represent their color distributions more efficiently than do histograms with fixed binnings. However, existing systems almost exclusively adopt fixed-binning histograms because, among existing well-known dissimilarity measures, only the computationally expensive Earth Mover's Distance (EMD) can compare histograms with different binnings. This article addresses the issue by defining a new dissimilarity measure that is more reliable than the Euclidean distance and yet computationally less expensive than EMD. Moreover, a mathematically sound definition of mean histogram can be defined for histogram clustering applications. Extensive test results show that adaptive histograms produce the best overall performance, in terms of good accuracy, small number of bins, no empty bin, and efficient computation, compared to existing methods for histogram retrieval, classification, and clustering tasks.

In this paper we introduce and describe the Multimedia Retrieval Markup Language (MRML). This XML-based markup language is the basis for an open communication protocol for content-based image retrieval systems (CBIRSs). MRML was initially designed as a means of separating CBIR engines from their user interfaces. It is, however, also extensible as the basis for standardised performance evaluation procedures. Such a tool is essential for the formulation and implementation of common benchmarks for CBIR. A common protocol can also bring new dynamics to the CBIR field -- it makes the development of new systems faster and more efficient, and opens the door of the CBIR research field to other disciplines such as Human-Computer Interaction. The MRML specifications, as well as the first MRML-compliant applications, are freely available and are introduced in this paper.

While in the area of relational databases interoperability is ensured by common communication protocols (e.g. ODBC/JDBC using SQL), Content Based Image Retrieval Systems (CBIRS) and other multimedia retrieval systems are lacking both a common query language and a common communication protocol. Besides its obvious short term convenience, interoperability of systems is crucial for the exchange and analysis of user data. In this paper, we present and describe an extensible XML-based query markup language, called MRML (Multimedia Retrieval Markup Language). MRML is primarily designed so as to ensure interoperability between dierent content{based multimedia retrieval systems. Further, MRML allows researchers to preserve their freedom in extending their system as needed. MRML encapsulates multimedia queries in a way that enables multimedia (MM) query languages, MM content descriptions, MM query engines, and MM user interfaces to grow independently from each other, reaching a maximum of in...

Histograms are commonly used in content-based image retrieval systems to represent the distributions of colors in images. It is a common understanding that histograms that adapt to images can represent their color distributions more efficiently than do histograms with fixed binnings. However, existing systems almost exclusively adopt fixed-binning histograms because among existing well-known dissimilarity measures, only the computationally expensive Earth Mover's Distance (EMD) can compare histograms with different binnings.

There is a high demand for effective and precise tools for users to search, browse, and interact with image databases and do so in a timely manner. Feature extraction is a crucial part for any such retrieval systems. Current methods for feature extraction suffer from two main problems: first, many methods do not retain any spatial information, and second, the problem of invariance with respect to standard transformation is still an unsolved problem. On the other hand, wavelets has been shown to be a powerful and efficient mathematical tool to process visual information at multiple scales. Some recent image retrieval systems use spatial information and visual features represented by dominant wavelet coefficients [1]. In addition, the underlying multiresolution mechanism of the wavelet decomposition allows the retrieval process to be done progressively. In this work, to cure the plague problem of translation variance with wavelet basis transform while keeping a compact representation, the wavelet transform modulus maxima [2] is employed. Wavelet maxima has been shown to be very effective in characterization of images from multiscale edges. Therefore feature extraction based on the wavelet maxima transform captures well the edge-based and spatial layout information which are likely the key features on an image query. To measure the similarity between wavelet maxima representations, which is required on the context of image retrieval systems, the difference of moments is used. Normalized central moments are efficiently computed for each scale of the wavelet maxima transform. As a result, each image is indexed by multiscale vectors in feature spaces. Moreover, those moments are invariant with respect to both translation and scaling. Evaluation of the proposed method was p...

This thesis describes research into an area of content based image retrieval (CBIR), that of feature indexing for the purpose of rapid retrieval. The techniques in this thesis draw from the field of text IR and demonstrate that individual image extraction algorithms can be optimised for use with an inverted index, which could lead to CBIR systems capable of sub-second retrieval times on collections of millions of images. A novel global feature algorithm...