Merih Seran Uysal

Signature Quadratic Form Distance

The Signature Quadratic Form Distance is an adaptive similarity measure for flexible content-based feature representations of multimedia data. In this paper, we present a deep survey of the mathematical foundation of this similarity measure which encompasses the classic Quadratic Form Distance defined only for the comparison between two feature histograms of the same length and structure. Moreover, we give the benefits of the Signature Quadratic Form Distance and experimental evaluation on numerous real-world databases.

A comparative study of similarity measures for content-based multimedia retrieval

Determining similarities among data objects is a core task of content-based multimedia retrieval systems. Approximating data object contents via flexible feature representations, such as feature signatures, multimedia retrieval systems frequently determine similarities among data objects by applying distance functions. In this paper, we compare major state-of-the-art similarity measures applicable to flexible feature signatures with respect to their qualities of effectiveness and efficiency. Furthermore, we study the behavior of the similarity measures by discussing their properties. Our findings can be used in guiding the development of content-based retrieval applications for numerous domains.

Signature quadratic form distances for content-based similarity

Determining similarity is a fundamental task in querying multimedia databases in a content-based way. For this challenging task, there exist numerous similarity models which measure the similarity among objects by using their contents. In order to cope with voluminous multimedia data, similarity models are supposed to be both effective and efficient. To this end, we introduce the Signature Quadratic Form Distance measure which allows efficient similarity computations based on flexible feature representations. Our new approach bridges the gap between the well-known concept of Quadratic Form Distances and feature signatures. Experimentation indicates that our similarity measure is able to compete with state-of-the-art similarity models regarding effectiveness of content-based similarity search. Moreover, our Signature Quadratic Form Distance outperforms the established Earth Movers Distance in efficiency: we obtain a speed-up factor of greater than 50.

Efficient Filter Approximation Using the Earth Mover's Distance in Very Large Multimedia Databases with Feature Signatures

The Earth Movers Distance, proposed in computer vision as a distance-based similarity model reflecting the human perceptual similarity, has been widely utilized in numerous domains for similarity search applicable on both feature histograms and signatures. While efficiency improvement methods towards the Earth Movers Distance were frequently investigated on feature histograms, not much work is known to study this similarity model on feature signatures denoting object-specific feature representations. Given a very large multimedia database of features signatures, how can k-nearest-neighbor queries be processed efficiently by using the Earth Movers Distance? In this paper, we propose an efficient filter approximation technique to lower bound the Earth Movers Distance on feature signatures by restricting the number of earth flows locally. Extensive experiments on real world data indicate the high efficiency of the proposal, attaining order-of-magnitude query processing time cost reduction for high dimensional feature signatures.

Endoscopic Video Retrieval: A Signature-Based Approach for Linking Endoscopic Images with Video Segments

In the field of medical endoscopy more and more surgeons are changing over to record and store videos of their endoscopic procedures, such as surgeries and examinations, in long-term video archives. In order to support surgeons in accessing these endoscopic video archives in a content-based way, we propose a simple yet effective signature-based approach: the Signature Matching Distance based on adaptive-binning feature signatures. The proposed distance-based similarity model facilitates an adaptive representation of the visual properties of endoscopic images and allows for matching these properties efficiently. We conduct an extensive performance analysis with respect to the task of linking specific endoscopic images with video segments and show the high efficacy of our approach. We are able to link more than 88% of the endoscopic images to their corresponding correct video segments, which improves the current state of the art by one order of magnitude.

Efficient similarity search in scientific databases with feature signatures

The recent rapid growth of scientific data necessitates efficient similarity search techniques for which convenient object representation models are of vital importance. Feature signatures denoting highly flexible object feature representations have increasingly gained attention for which corresponding efficiency improvement techniques are developed. In this paper, we focus on efficient query processing with the well-known Earth Movers Distance (EMD) on databases of feature signatures, and propose efficient approximation techniques successfully applicable to high-dimensional feature signatures via dimensionality reduction, guaranteeing both completeness and no false-dismissal within a filter-and-refine architecture. Rigorous experiments on real world data indicate a considerable reduction in the number of EMD computations and high efficiency of the proposed techniques which significantly reduce the query processing time.

Efficient k-nearest neighbor queries with the Signature Quadratic Form Distance

A frequently encountered query type in multimedia databases is the k-nearest neighbor query which finds the k-nearest neighbors of a given query. To speed up such queries and to meet the user requirements in low response time, approximation techniques play an important role. In this paper, we present an efficient approximation technique applicable to distance measures defined over flexible feature representations, i.e. feature signatures. We apply our approximation technique to the recently proposed Signature Quadratic Form Distance applicable to feature signatures. We performed our experiments on numerous image databases, gathering k-nearest neighbor query rankings in significantly low computation time with an average speed-up factor of 13.

Content-based exploration of multimedia databases

Usability, effectiveness, and efficiency are the fundamental properties of content-based multimedia exploration systems. While the usability of an exploration system is frequently ensured by intuitive and interactive visual interfaces, the effectiveness of the retrieval results and the efficiency of the exploration process are typically obtained by high quality similarity models and fast query evaluation strategies. As it is immensely important to maintain all the aforementioned properties concurrently in order to meet individual user requirements, we present a modular content-based multimedia exploration system facilitating the access and insight into large multimedia databases by incorporating state-of-the-art content-based methods and techniques. To this end, we first survey and analyze various techniques and then briefly introduce our modular content-based exploration system.

On efficient content-based near-duplicate video detection

The high usage of the Internet, in particular videosharing and social networking websites, have led to enormous amount of video data recently, raising demand on effective and efficient content-based near-duplicate video detection approaches. In this paper, we propose to efficiently search for near-duplicate videos via the utilization of efficient approximation techniques of the well-known effective similarity measure Earth Movers Distance (EMD). To this end, we model keyframes by flexible feature representations which are then exploited in a filter-and-refine architecture to alleviate the query processing time. The experiments on real data indicate high efficiency guaranteeing reduced number of EMD computations, which contributes to the near-duplicate detection in video datasets.

Similarity matrix compression for efficient signature quadratic form distance computation

Determining similarities among multimedia objects is a fundamental task in many content-based retrieval, analysis, mining, and exploration applications. Among state-of-the-art similarity measures, the Signature Quadratic Form Distance has shown good applicability and high quality in comparing flexible feature representations. In order to improve the efficiency of the Signature Quadratic Form Distance, we propose the similarity matrix compression approach which aims at compressing the distances inherent similarity matrix. We theoretically show how to reduce the complexity of distance computations and benchmark computation time improvements. As a result, we improve the efficiency of a single distance computation by a factor up to 9.