A Flexible Image Retrieval Framework

Abstract

Abstract. This paper discusses a framework for image retrieval. Most current systems are based on a single technique for feature extraction and similarity search. Each technique has its advantages and drawbacks concerning the result quality. Usually they cover one or two certain features of the image, e.g. histograms or shape information. The proposed framework is designed to be highly flexible, even if performance may suffer. The aim is to give people a platform to implement almost any kind of retrieval issues very quickly, whether it is content based or somehing else. The second advantage of the framework is the possibility to change retrieval characteristics within the program completely. This allows users to configure the ranking process as needed.

An extensible query language for content based image retrieval based on Lucene

“…The plug in concept of the retrieval framework described in [15] allows the definition of any new feature. To make such a plug in available in this language, only a few requirements need to be met.…”

“…The language presented in this paper represents the middleware of the previously described retrieval framework [15]. It is easily parseable and allows composition of any query that is supported by the framework.…”

Using CBIR and semantics in 3D-model retrieval

“…Many parts can be constructed very similar to the underlying CBIR framework [20]. For this reason only the differences are emphasized.…”

“…The prototype design ( fig. 2) follows the previously published framework [20] which is responsible to store and manage the repository.…”

“…Query composition, browsing and the result representation require techniques aligned to 3D specifics. This paper discusses a possible design for a 3DMR program based on a previously published CBIR framework [20], the EADS demonstrator [15] and a newly developed query language [? ].…”