Navidgator-similarity based browsing for image and video databases

Proceedings of the 1st international workshop on Computer vision meets databases - CVDB '04, 2004

Browsing large multimedia databases is becoming a challenging problem, due to the availability of great amounts of data and the complexity of retrieval. In this paper we propose a system that assists a user in browsing a digital collection making useful recommendations. The system combines computer vision techniques and taxonomic classifcations to measure the similarity between objects and adopts an

2004

2004

In this puper, rhe asthorr propose an interactive grapLica1 interjace suitable for semantic browsing of an image database. This interface allows user 10 explore the eflects of features integration. multi-image queries and relevonce feedback in enhancing the performance ofan image refrieval system. Single image query is the retrieval model that is traditionally used, while the outhors propose the implementation of a new methodology consisting of multiple query images. Features integration and feedback improve the performance ofrhe proposed system signi/icnnrly. These techniques require an interactive interfce which is able to gather information from user and to show him the view of the image database closest to his requirements.

IEEE Multimedia, 2000

The image database system El Niño uses a new interaction model that aims to overcome the problem of the semantic gap where the meaning that the user has in mind for an image is at a higher semantic level than the features on which the database operates. To solve this problem, we replaced the usual query paradigm with a more active exploration process and developed an interface based on these premises.

Proceedings of the 16th IFIP World …, 2000

Digital images and videos have an increasingly important role in today's telecommunication and our everyday life in modern information society. The past few years witnessed a proliferation of content-based image retrieval techniques. Images are typically characterized by intrinsic attributes of images such as color, texture, and shape. However, the potential of integrating these techniques with visualization and data-mining techniques has yet been fully explored. Users should be able to explore images in a database or video clips by visual similarities. In this article, we explore the synergy between Pathfinder networks and content-based information retrieval techniques. Salient structures of images are revealed through visualization models derived from features extracted from images. Visualizations are generated from three feature classes of the well-known QBIC system: color, layout, and texture.

This paper presents the main features of a Multimedia Query Language tailored for content-based similarity retrieval of multimedia objects. The Query Language processor is a component of a multimedia database system that adopts a model that permits both a structural representation of raw multimedia data and an automatically computed description of the multimedia data content. The Query Language is an extension of a traditional object-oriented query language. It allows to express restrictions on features, concepts and structural aspects of the multimedia database objects. In addition, the language supports the formulation of queries with imprecise conditions. The outcome of a query execution is an ordered set of pairs, each one consisting of an object and a measure of the similarity of the object with the criteria specified in the query.

2013

VideoCycle is a candidate application for this second Video Browser Showdown challenge. VideoCycle allows interactive intra-video and inter-shot navigation with dedicated gestural controllers. MediaCycle, the framework it is built upon, provides media organization by similarity, with a modular architecture enabling most of its workflow to be performed by plugins: feature extraction, clustering, segmentation, summarization, intra-media and inter-segment visualization. MediaCycle focuses on user experience with user interfaces that can be tailored to specific use cases.

Key Technologies for Data Management

In order to improve the robustness of systems that perform similarity searching, it may be necessary to augment a collection of images in a Multimedia DataBase Management System (MMDBMS) with an additional set of edited images. Previous research has demonstrated that space can be saved in such a system by storing the edited images as sequences of editing operations instead of as large binary objects. The existing approaches for performing similarity searching in an MMDBMS, however, typically assume that the data objects are stored as binary objects. This paper proposes algorithms for performing color-based similarity searches of images stored as editing operations and provides a performance evaluation illustrating their respective strengths and weaknesses.

2009

The state of the art of searching for non-text data (e.g., images) is to use extracted metadata annotations or text, which might be available as a related information. However, supporting real content-based audio-visual search, based on similarity search on features, is significantly more expensive than searching for text. Moreover, such search exhibits linear scalability with respect to the data set size, so parallel query execution is needed.