[Show abstract][Hide abstract] ABSTRACT: In this paper, we propose an extended framework structure designed for MUVIS multimedia indexing and retrieval scheme in order to achieve the dynamic integration and run-time execution for the following operations within the context of multimedia indexing and retrieval: Visual and aural feature extraction, shot boundary detection and spatial segmentation. We introduce a general structure of the extended framework, and describe its well-defined interface, the general-purpose instructions and the conditions for integrating new algorithms into MUVIS. Finally, the limitations and the advantages of the integration scheme are discussed whilst providing examples from some algorithms already implemented.
[Show abstract][Hide abstract] ABSTRACT: This paper presents a visual media querying scheme referred to as transform-based layered query (TLQ) scheme. The TLQ scheme mainly aims at decreasing retrieval processing time and run-time memory consumption without degrading retrieval results semantically. The scheme contains abstract layers in indexing and retrieval phases, where each indexing layer corresponds to a retrieval layer. The layers are constructed based on transformations for reducing visual frame and feature data dimensions. The proposed TLQ scheme also involves an unsupervised method for eliminating irrelevant media items between the retrieval layers. A two-layer TLQ system is implemented and integrated into MUVIS content-based multimedia indexing and retrieval framework, and its theoretical advantages are verified with dedicated experiments on image and video databases. The experiments reveal that 75% retrieval performance improvement in terms of process time can be achieved depending on transformation parameters.
[Show abstract][Hide abstract] ABSTRACT: This paper presents a content-based multimedia indexing and retrieval framework designed for mobile platforms run-ning Symbian-based operating system. It is mainly devel-oped by focusing on mobile platform restrictions and Symbian features. The proposed framework is built on a client-server architecture, where the client side basically consists of the user interface and controllers, and the server is responsible for performing main functions. Both the client and the server may run on either the same mobile device, or separately on connected devices. A sample content-based image indexing and retrieval application running on single Nokia 6630 device is implemented for testing the frame-work, verify its feasibility, and study further directions in the area. The implemented application provides mobile indexing and retrieval features as well as instant image capturing with onboard camera for queries. The experimental studies reveal relatively successful results in terms of semantic and process performance.
[Show abstract][Hide abstract] ABSTRACT: To employ an existing software library, its structure should be first learned and the required elements should be identified. This can be challenging if the library is large and only a specific part of it should be comprehended. In this paper, we study the problem of learning complex software libraries modeled in UML. It is argued that the learning process can be supported with a tool environment that allows the customization of the UML model according to the context of the learner, stepwise and dynamically chosen learning tasks, and focusing on a particular learning concern at a time. We show how such an environment can be achieved based on the concept of a pattern, using existing tool support. We demonstrate the idea with a part of Symbian platform architecture. The approach is evaluated in a case study where a pattern-driven learning environment is constructed for JPEG interchange file format specifications.
[Show abstract][Hide abstract] ABSTRACT: This paper presents an evaluation of digital compression effects on content-based multimedia retrieval using color and texture attributes. Subjective evaluation tests that are applied on digital image and video databases using different compression and visual feature extraction techniques have been performed and reported. Simulations show that a satisfactory retrieval performance can be obtained from the compressed databases with 10% compression quality (i.e. 97.6% compression ratio in JPEG). Image retrieval based on HSV color histogram performs better than retrieval based on YUV color histogram in the uncompressed domain, and the other way around in the compressed domain. In general, video retrieval based on color histogram in MPEG-4 compressed databases performs better compared to H.263+ compressed databases. However, retrieval performance from H.263+ compressed databases at lower bit rates is more stable, where it drastically decreases in MPEG-4 compressed databases below 128 Kb/s. Retrieval based on texture features produces more robust performance than retrieval based on color. Subjective tests show that 25% compression quality achieves high compression ratio without loosing significant retrieval performance. The results are particularly relevant to applications in which a mobile device is involved in a multimedia retrieval system.
[Show abstract][Hide abstract] ABSTRACT: MUVIS is a series of CBIR systems. The first one has been developed in late 90s to support indexing and retrieval in large image databases using visual and semantic features such as color, texture and shape. During recent years, MUVIS has been reformed to become a PC-based framework, which supports indexing, browsing and querying of various multimedia types such as audio, video, audio/video interlaced and several image formats. MUVIS system allows real-time audio and video capturing, encoding by last generation codecs such as MPEG-4, H.263+, MP3 and AAC. It supports several audio/video file format such as AVI, MP4, MP3 and AAC. Furthermore, MUVIS system provides a well-defined interface for third parties to integrate their own feature extraction algorithms into the framework and for this reason it has recently been adopted by COST 211quat as COST framework for CBIR. In this paper, we describe the general system features with underlying applications and outline the main philosophy.
[Show abstract][Hide abstract] ABSTRACT: In this paper, we present an integration scheme de-signed in a content-based multimedia indexing and re-trieval framework (MUVIS) for independent feature ex-traction algorithms. We introduce the general structure of the framework, and describe the interface, the instruc-tions and the conditions for integrating a feature extrac-tion algorithm into MUVIS. Finally, the limitations and the advantages of the integration scheme are discussed by giving examples from the implemented algorithms.
[Show abstract][Hide abstract] ABSTRACT: This paper evaluates the performance of texture and color features applied to natural image retrieval. Natural images usually contain both texture and color, and therefore those features should also be considered together in a retrieval process. Combination of texture and color features is important, since it may result in closer connection to semantic. The idea in this study is to empirically evaluate the effect of texture and color features to the query. Retrieval capability of those features is first evaluated separately and finally the combined effect is studied. Finally, the goal is to find out what kind of features, or combination of features, yield the most successful results for certain types of query images.
[Show abstract][Hide abstract] ABSTRACT: First MUVIS system has been developed three years ago, supporting image indexing and means to retrieve images from large image databases using image visual and semantic features such as color, texture and shape. Recently, MUVIS project has been reformed to become a PC-based system, which supports indexing, browsing and querying on various multimedia types such as audio, video and image. Furthermore the system allows real-time audio and video capturing, encoding by last generation codecs such as MPEG-4 or H.263+ if requested, recording while indexing into a database in such a way that they can be retrieved efficiently. In this paper, we describe the system features with underlying applications and outline the mean philosophy. Query and browsing capabilities of the MUVIS technology will be demonstrated during the conference.
[Show abstract][Hide abstract] ABSTRACT: MUVIS is a PC-based framework, which supports indexing, browsing and querying of various multimedia types such as audio, video, audio/video interlaced in several formats. It allows real-time audio and video capturing, encoding by last generation codecs such as MPEG-4, H.263+, MP3 and AAC. MUVIS also supports several audio/video file format such as AVI, MP4, MP3 and AAC. Almost all image types in a PC environment including JPEG-2000 can be rendered, indexed and converted within MUVIS framework. Along with such a wide multimedia coverage, MUVIS has been developed to achieve a global and unified solution for content-based indexing and retrieval problem and to provide user-friendly applications and a generic framework especially for third parties to develop their feature extraction modules. In this paper, we present an overview of the MUVIS system and we shall especially focus on the overall audio-based multimedia indexing and retrieval scheme within MUVIS framework.
[Show abstract][Hide abstract] ABSTRACT: Content-based image indexing complexity often depends on image dimensions and data size. Reducing image dimensions before indexing affects the overall feature extraction and indexing complexity, as well as the retrieval performance. In this paper, the effects of Discrete Wavelet Transform (DWT) based downscaling on semantic retrieval performance are investigated via dedicated experiments. Several images are collected from various sources and experimental databases are generated by applying JPEG2000 compression and DWT-based downscaling. Certain images are queried through the databases based on their color, texture, and shape features in the retrieval performance experiments. The evaluation results show that DWT-based downscaling does not have significant impact on color and texture-based retrieval in general, while it degrades edge-based retrieval performance significantly. The results also verify that JPEG2000 compression does not affect texture and edge-based retrieval performance, although it decreases the color-based image retrieval performance.