H. J. Zhang, J. Y. A. Wang, and Y. Altunbasak, "Content-based video retrieval and compression: A unified solution," Proc. IEEE Int. Conf. Image Processing, vol. 1, pp. 13--16, 1997.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....of video; 2) video classification; and 3) extraction of semantics. The work in the first category has concentrated on: 1) shot boundary detection and key frame extraction, e.g. 1] 34] 2) shot clustering, e.g. 32] 3) table of content creation, e.g. 9] 4) video summarization, e.g. 22] [35]; and 5) video skimming [28] Manuscript received December 15, 1998; revised December 28, 1999. This paper was recommended by Associate Editor S. F. Chang. N. Haering is with Diamondback Vision, Inc. Reston, VA 20191 USA (e mail: niels dbvision.net) R. J. Qian is with Intel Labs, Hillsboro, OR ....

H. J. Zhang, J. Y. A. Wang, and Y. Altunbasak, "Content-based video retrieval and compression: A unified solution," Proc. IEEE Int. Conf. Image Processing, vol. 1, pp. 13--16, 1997.

....the multijects using a multinet in the form of a Bayesian network. Experiments reveal significant performance improvement using the multinet. 1. Introduction Research in video indexing and retrieval has traditionally focussed on the paradigm of query by example (QBE) Related work includes [14] [13] etc. The assumption of availability of clips that can be used as examples is very demanding. The ability to query with key words or key phrases (preferably semantic) instead of examples has motivated recent research in semantic video indexing. To facilitate this it is necessary to have models ....

H. Zhang, A. Wang, and Y. Altunbasak. Content-based video retrieval and compression: A unified solution. In Proceedings of the IEEE International Conference on Image Processing, volume 1, pages 13--16, Santa Barbara, CA, Oct. 1997.

....recent work in the field of semantic indexing and present our statistical framework for automatic indexing. 3 Semantic Video Indexing: A novel framework Much of the work in domain independent video retrieval has focussed on the Query by example paradigm. Examples of such system can be found in [18, 1, 17, 10]. A Query by example (QBE) system matches user query with the targets in a database. It is implicitly assumed that the form of the query must be identical to that of the target. To query a video database by example, one must therefore be able to frame the query in terms of a suitable video clip. ....

H. Zhang, A. Wang, and Y. Altunbasak. Content-based video retrieval and compression: A unified solution. In Proceedings of the IEEE International Conference on Image Processing, volume 1, pages 13--16, Santa Barbara, CA, Oct. 1997.

....explosions and waterfalls. 1. INTRODUCTION Multimedia Indexing and Retrieval presents a challenging task of developing algorithms that fuse information from multiple media to support queries. The state of the art in content based video indexing and retrieval techniques includes [1] 2] and [3]. While [3] is oriented towards effective browsing support, 1] uses various features like color texture, shape and motion for video indexing and retrieval. There has been work in This research was supported by the NCSA ISAAC project and a grant from the Arnold and Mabel A. Beckman foundation. ....

....and waterfalls. 1. INTRODUCTION Multimedia Indexing and Retrieval presents a challenging task of developing algorithms that fuse information from multiple media to support queries. The state of the art in content based video indexing and retrieval techniques includes [1] 2] and [3] While [3] is oriented towards effective browsing support, 1] uses various features like color texture, shape and motion for video indexing and retrieval. There has been work in This research was supported by the NCSA ISAAC project and a grant from the Arnold and Mabel A. Beckman foundation. Milind R. ....

[Article contains additional citation context not shown here]

H. Zhang, J. A. Wang and Y. Altunbasak, "Content-Based Video Retrieval and Compression: A Unified Solution", IEEE Intl. Conf. on Image Processing, Vol 1 pp. 13-16, Oct 1997, Santa Barbara ,CA.

....relation with observable concepts. Our experiments reveal significant performance improvement using the inference on the factor graph models. 1. Introduction The most prominent research in video indexing and retrieval has traditionally focussed on the paradigm of queryby example (QBE) 12] [11], 2] Query by keywords (QBK) or key phrases (preferably semantic) instead of examples has motivated recent research in semantic video indexing. For this we need models which capture the representation corresponding to these keywords. A QBK system can support semantic retrieval using a ....

H. Zhang, A. Wang, and Y. Altunbasak. Content-based video retrieval and compression: A unified solution. In Proceedings of the IEEE International Conference on Image Processing, volume 1, pages 13--16, Santa Barbara, CA, Oct. 1997.

....the multijects using multinet in the form of a Bayesian network. Detection performance is significantly improved using the multinet. 1. Introduction Research in video indexing and retrieval has traditionally focussed on the paradigm of query by example (QBE) Related work includes [14] [13] etc. The assumption of availability of clips that can be used as examples is demanding. The ability to query with key words or key phrases (preferably semantic) instead of examples has motivated recent research in semantic video indexing. To facilitate this it is necessary to have models which ....

H. Zhang, A. Wang, and Y. Altunbasak. Content-based video retrieval and compression: A unified solution. In Proceedings of the IEEE International Conference on Image Processing, volume 1, pages 13--16, Santa Barbara, CA, Oct. 1997.

....have explosions and waterfalls. 1. INTRODUCTION Multimedia Indexing and Retrieval presents a challenging task of developing algorithms that fuse information from multiple media to support queries. The state of the art in content based video indexing and retrieval techniques includes [1] 2] [3]. While [3] is oriented towards effective browsing support [1] uses various features like color texture, shape and motion for video indexing and retrieval. There has been work in indexing for specific domains [4] 5] limiting the scenario This research was supported by the NCSA ISAAC project and ....

....and waterfalls. 1. INTRODUCTION Multimedia Indexing and Retrieval presents a challenging task of developing algorithms that fuse information from multiple media to support queries. The state of the art in content based video indexing and retrieval techniques includes [1] 2] 3] While [3] is oriented towards effective browsing support [1] uses various features like color texture, shape and motion for video indexing and retrieval. There has been work in indexing for specific domains [4] 5] limiting the scenario This research was supported by the NCSA ISAAC project and a grant ....

[Article contains additional citation context not shown here]

H. Zhang, J. A. Wang and Y. Altunbasak, "Content-Based Video Retrieval and Compression: A Unified Solution", IEEE Intl. Conf. on Image Processing, Vol 1 pp. 13-16, Oct 1997, Santa Barbara ,CA.

....grouping, and scene structure construction. We discuss each of the modules in turn below: 3.1 Shot Boundary Detection and Key Frame Extraction As described in Section 2. 1, a lot of work has been done in shot boundary detection and many of the approaches achieve satisfactory performance [22]. In this paper, we use an approach similar to the one used in [23] For key frame extraction, although more sophisticated techniques exist [27, 8, 15] they require high computation effort. We select the beginning and ending frames of a shot as the two key frames to achieve fast processing speed. ....

H.J. Zhang, John Y. A. Wang, and Yucel Altunbasak. Content-based video retrieval and compression: A unified solution. In Proc. IEEE Int. Conf. on Image Proc., 1997.

....when they enjoy video content. The objectbased video retrieval is a much better method for users [4] However, the semantic object segmentation still needs for human s interaction at the current stage. Also, in the most of previous works, the visual features were extracted based on the key frame [5]. Such representation may not be complete due to the spars nature of key frames, especially for motions in video. To overcome this shortcoming, some shot based features extraction methods have been proposed for color description recently [6] 7] In this paper, a shot based motion energy ....

H.J.Zhang,et al, "Content-based Video Retrieval and Compression: A Unified Solution", proceeding of ICIP'97, pp.13-16, 1997.

No context found.

H. Zhang, A. Wang, and Y. Altunbasak, "Content-based video retrieval and compression: A unified solution," in Proceedings of the IEEE International Conference on Image Processing, vol. 1, pp. 13--16, (Santa Barbara, CA), Oct. 1997.

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