Philippe Aigrain, HongJiang Zhang and Dragutin Petkovi "Contentbased Representation and Retrieval of Visual Media: A State-of-the-Art Review", Multimedia Tools and Applications, vol 3, no 3, November 1996, pp: 179-202.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....image. Superimposed text, small images and logos appearance and disappearance are not counted as transition e ects. Many automated tools for the temporal segmentation of video streams have been already proposed. It is possible to nd some papers that are providing state of the art of such methods [3] [4] 5] In this paper, we describe the temporal video segmentation system used by CLIPS IMAG to perform the Shot Boundary Detection (SBD) task of the Video track of the TREC 10 conference. This system was rst developed at the LIMSI CNRS laboratory and was then improved at the CLIPS IMAG ....

Aigrain, P., Zhang, H.J., Petkovic, D.,: Content-based representation and retrieval of visual media : a state-of-the-art review, In Multimedia Tools and Applications, 3(3):179-202, November 1996

....related to their content to satisfy a given query or to access particular pieces of information. There is obviously a real need of indexing and retrieving multimedia documents by their content in an automatic way (at least partly) Several pioneering systems already exist for still images, [1, 5], and a large research effort is currently undertaken to handle image and video databases, 1, 7, 9, 13, 16] Nevertheless, due to the complexity of image interpretation and dynamic scene analysis, several important issues remain to be further investigated. As far as video sequences are ....

.... There is obviously a real need of indexing and retrieving multimedia documents by their content in an automatic way (at least partly) Several pioneering systems already exist for still images, 1, 5] and a large research effort is currently undertaken to handle image and video databases, [1, 7, 9, 13, 16]. Nevertheless, due to the complexity of image interpretation and dynamic scene analysis, several important issues remain to be further investigated. As far as video sequences are concerned, contentbased video indexing, browsing, editing, or retrieval, have motivated specific investigations ....

[Article contains additional citation context not shown here]

P. Aigrain, H. J. Zhang, and D. Petkovic. Content-based representation and retrieval of visual media : A stateof -the-art review. Multimedia Tools and Applications, 3(3):179-202, Nov. 1996.

.... for segmenting video into groups of frames is through the popular shot based representation model [1] 2] Once the shot boundaries in a video sequence are identified, it is customary to describe the visual and color content of shots using key frames and key frame histograms, respectively [3]. Although the key frame color histogram is a very simple and, depending on how the key frames are chosen, computationally inexpensive descriptor of color content of a shot, the color 3 description it provides varies significantly with the selection criterion. Some methods simply pick from every ....

....be directly extended to collections of still images, as long as a meaningful grouping or clustering of the images in a database is provided. A. Key Frame Histogram Key frame histogram refers to the color histogram of a representative frame of a GoF that is selected according to some criterion [3], 9] Although this approach is very simple and (often) computationally the least expensive, the color description provided by the resulting histogram varies significantly with the key frame selection criterion. In the following, we propose a particular key frame selection method to avoid this ....

P. Aigrain, H. J. Zhang, and D. Petkovic, "Content-based representation and retrieval of visual media: A state-of-the-art review," Multimed. Tools Appl., vol. 3, no. 3, pp. 3-26, 1996.

....confidential algorithms to compress multi dimensional query document relationship information (section 2.8.1) into a linear list. These algorithms are not well understood by users, particularly algorithms that incorporate different types of evidence, e.g. a combination of text and content analysis [2, 34, 28]. Control: Inexpressive Query Language The large number of images indexed by WWW image retrieval engines makes content based image analysis techniques (section 3.3.2) difficult to apply. Advanced image analysis techniques are computationally expensive to run. Further, the effectiveness ....

....algorithms declines when used over a collection with a large breadth of content [56] Existing infrastructure used by WWW search engines to perform image retrieval provides a limited capacity for users to specify their precise image needs. Current systems allow only for text based image queries [2, 28]. In providing a search service over a high latency network, current WWW image retrieval systems are limited to providing coarse grained interaction. In current systems, users must submit a query, retrieve results and then choose either to restate the query or perform a find similar search. ....

[Article contains additional citation context not shown here]

AIGRAIN, P., ZHANG, H., AND PETKOVIC, D. Content-based representation and retrieval of visual media: A state-of-the-art review. In Multimedia tools and applications (1996), vol. 3, pp. 179--202.

....number of key frames to match low level browsing preferences of a user. The proposed method has been validated by experimental results on a collection of video programs. I. INTRODUCTION A number of methods have been proposed in the literature to generate visual summaries of video programs [1] [2], 3] 4] 5] 6] Recently, there has been renewed interest in improved techniques in the context of MPEG 7 standardization. MPEG 7 (formally Multimedia Con tent Description Interface ) is an emerging international standard for syntactic and semantic description of multimedia content. It ....

....respectively. Experimental results that demonstrate the effectiveness of the method are provided in Section 5. II. A ROBUST DESCRIPTOR FOR MULTIFRAME COLOR REPRESENTATION A common way to describe the color content of a video shot or segment is to use the histograms of one or more of its frames [2]. Although this approach provides a simple and often computa tionally inexpensive color descriptor, the quality of the resulting description varies significantly with how the representative frames are selected. As a robust alternative, the alpha trimmed average histogram, Ha s , was recently ....

P. Aigrain, H. J. Zhang, and D. Petkovic, "Content-based representation and retrieval of visual media: A state-of-the-art review," Multimedia Tools and Applications: Sp. Isssue on Rep. and Ret. of Visual Media in Multimedia Systems, vol. 3, no. 3, pp. 3 26, 1996.

....These developments can be seen as origins of multimedia analysis, now in its relative embryonic infancy. What might be seen as a progression of the previous image work is today s developments in automatic classification and annotation of multimedia material. Reviews are presented by Aigrain et al. [4] and more recently Brunelli et al. [5] Also Wang et al. [6] presented an account with an acoustic bias. There are many different aspects and approaches to content based multimedia analysis, the end application being an important determining factor. Motivation for contentbased multimedia analysis ....

....key frames, in a mosaic form, or a kind of importance measures etc, the latter is almost certain to be in the form of a shorter semantics preserved video file. See for example [14] In this sense, one of the most important desired applications is real time adaptive video streaming . Aigrain et al. [4] suggest in their paper that video skimming is more successful in limited domains such as education or news videos; namely those with very explicit speech or text (closed caption) content. 2.3 Video indexing or labelling Video indexing differs from retrieval by example in that video indexing ....

P. Aigrain, H. Zhang, and D. Petkovic, "Content-based representation and retrieval of visual media: A state of the art review, " Multimedia Tools and Applications, vol. 3, pp. 179--202, 1996.

....These developments can be seen as origins of multimedia analysis, now in its relative embryonic infancy. What might be seen as a progression of the previous image work is today s developments in automatic classification and annotation of multimedia material. Reviews are presented by Aigrain et al. [4] and more recently Brunelli et al. [5] Also Wang et al. [6] presented an account with an acoustic bias. There are many different aspects and approaches to contentbased multimedia analysis, the end application being an important determining factor. Motivation for content based multimedia analysis ....

....key frames, in a mosaic form, or a kind of importance measures etc, the latter is almost certain to be in the form of a shorter semantics preserved video file. See for example [14] In this sense, one of the most important desired applications is real time adaptive video streaming . Aigrain et al. [4] suggest in their paper that video skimming is more successful in limited domains such as education or news videos; namely those with very explicit speech or text (closed caption) contents. 2.3. Video indexing or labelling This differs from retrieval by example in that these approaches attempt ....

P. Aigrain, H. Zhang, and D. Petkovic, "Content-based representation and retrieval of visual media: A state of the art review," Multimedia Tools and Applications, vol. 3, pp. 179--202, 1996.

....obtain and to store in a database, and their matching takes little time. Inevitably, the global methods lack the power of locating specific objects and identifying their details (size, position, orientation, etc. Some extensions to the global method include search by color layout [2] by sketch [6, 9], and by color regions according to their spatial arrangements [7, 8] It has long been argued that segmentation plays an important role in human vision [19] Ever since the 1960s and 1970s, image segmentation has been one of the most persistent research areas in computer vision. It is hoped that ....

....Engine) The Image Excavator extracts images from an image repository. This repository can be the WWW space; in such case, the process crawls the Web searching for images, or a set of still images on disk or CD ROM. Frames can also be extracted from video streams using cut detection algorithms [35, 18, 6] and processed as still images. Once images are extracted from the repository, they are given as input to the image analyzer (C BIRD preprocessor) that extracts visual content features like color and edge characteristics. These visual features, FIG. 2. a) C BIRD general architecture. b) ....

P. Aigrain, H. Zhang, and D. Petkovic, Content-based representation and retrieval of visual media: A stateof -the-art review, Int. J. Multimedia Tools Appl. 3, November 1996, 179--202.

....The corresponding encoded information is far too complex to be extracted automatically from the video input and requires tedious manual encoding. As a consequence, there is a real need for video structuring tools based on feature extraction and relying on visual content (see for instance [1, 2]) In this work, we consider this point of view with a particular emphasis on video browsing applications. We restrict the field to image sequence analysis and acknowledge that sound processing is a convenient additional tool for video processing, especially when considering applications such as ....

.... sequence analysis and acknowledge that sound processing is a convenient additional tool for video processing, especially when considering applications such as video conferences ( 18, 24] Video has led to its own direction in the research carried out for accessing content based information, [1, 2, 30]. Issues concerned with video content representation benefit from studies devoted to temporal video segmentation, mosaic image construction or object tracking, to build iconic shot summaries, 6, 7, 10, 25] and these were evaluated to be comfortable [5] Nevertheless, these schemes enable only to ....

P. Aigrain, H.-J. Zhang, and D. Petrovic. Content-based representation and retrieval of visual media. Multmedia Tools and Applications, 3:179--202, 1996.

....elds. Reliable and convenient access to visual information is then of major interest for an eOEcient use of these databases. This implies indexing and retrieval of visual documents by their content. A great deal of research amount is currently devoted to image and video database management, [AZP96,BMM99]. Nevertheless, it remains hard to easily identify relevant information with regards to a given query, due to the complexity of dynamic scene analysis. Another important aspect of video database management lies in the denition of appropriate similarity measures associated to the description of ....

Aigrain (P.), Zhang (H-J.) et Petkovic (D.). Content-based representation and retrieval of visual media : A state-of-the-art review. Multimedia Tools and Applications, vol. 3, n 3, September 1996, pp. 179202.

No context found.

P. Aigrain, H. J. Zhang and D. Petkovic, "Content-Based Representation and Retrieval of Visual Media: A State-of-the-Art Re view", Int. J. Multimedia Tools Appl., pp.179---202, Vol. 3, November 1996.

....image video analysis, audio and speech analysis, user interface, information retrieval and artificial intelligence. New Information Retrieval (IR) search and browse techniques are needed that take into account the large size and nature of media as well as errors inevitably present in indexing [2]. Most of research community is mainly focused in areas of news, sport, entertainment and government applications. Issues related to video for education and training tend to receive much less attention (for an overview of issues managing video assets and summary of emerging integrated video ....

H. Zhang, P. Aigrain and D. Petkovic, "Content-Based Representation and Retrieval of Visual Media: A State-of-the-Art Review", Multimedia Tools and Applications, Vol. 3, pp. 179-202, Kluwer Academic Publishers, 1996.

....or user interaction, while extraction of visual features can be often done automatically and it is usually domain independent. Extensive research efforts have been made with regard to the retrieval of video and image data based on their features such as colour distribution, texture and shape [3]. These approaches fall into two categories: query by example and visual sketches. Both of these are based on similarity measurement. Examples include IBM s Query by Image Content (QBIC) 4] VisualSEEk [5] Photobook [6] Blobworld [7] as well as Virage video engine [8] CueVideo [9] and VideoQ ....

P. Aigrain, H. Zhang, D. Petkovic, Content-based Representation and Retrieval of Visual Media: A State-ofthe -Art Review, Multimedia Tools and Applications, Kluwer Academic Publishers, 3(3), 1996, 179-202.

No context found.

Philippe Aigrain, HongJiang Zhang and Dragutin Petkovi "Contentbased Representation and Retrieval of Visual Media: A State-of-the-Art Review", Multimedia Tools and Applications, vol 3, no 3, November 1996, pp: 179-202.

No context found.

Philippe Aigrain, HongJiang Zhang, and Dragutin Petkovic. Content-based representation and retrieval of visual media: A state-of-the-art review. Multimedia Tools and Applications, 3(3):179-202, November 1996.

No context found.

Philippe Aigrain, HongJiang Zhang, and Dragutin Petkovic. Content-based representation and retrieval of visual media: A state-of-the-art review. Multimedia Tools and Applications, 3(3):179-202, November 1996.

No context found.

P. Aigrain, H. Zhang, and D. Petkovic. Content-based representation and retrieval of visual media: A state-of-the-art review. Multimedia Tools and Applications, 3(3):179--202, 1996.

No context found.

P. Aigrain, H.J. Zhang, and D. Petkovic. Content-based representation and retrieval of visual media : a state-of-the-art review. Multimedia Tools and Appl., 3:179--202, 1996.

No context found.

P. Aigrain, H.J. Zhang and D. Petkovic, Content-based representation and retrieval of visual media: a state of the art review, Multimedia Tools and Applications, vol. 3, pp.179-202, 1996.

No context found.

P. Aigrain, H. Zhang, and D. Petkovic, "Content-Based Representation and Retrieval of Visual Media: A State-of-theArt Review," Multimedia Tools and Applications, Vol. 3, 179-202, November 1996.

No context found.

P. Aigrain, H. Zhang , and D. Petkovic, Content-Based Representation and Retrieval of Visual Media: A State-of-the-art Review, Multimedia Tools and Applications, 3(3), pp 179-202, 1996.

No context found.

P. Aigrain et al., Content-based representation and retrieval of visual media -- a state-of-the-art review, Multimedia Tools and Applications, 13(3), 1996, 179-202.

No context found.

P. Aigrain, H-J. Zhang, and D. Petkovic. Content-based representation and retrieval of visual media : A state-of-the-art review. Multimedia Tools and Applications, 3(3):179--202, 1996.

No context found.

P. Agrain; H. Zhang; D. Petkovic; "Content-based representation and retrieval of visual media: a state of the art review", Multimedia Tools and Applications, 3(3), pp. 179-202, 1997.

No context found.

Aigrain, P., Zhang, H.J.,Petkovic, D.: Content-based Representation and Retrieval of Visual Media: A State-of-the-Art Review. Multimedia Tools and Applications, 3(3):179--202, July 1996.

First 50 documents  Next 50

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC