Yong Rui, Thomas S. Huang and Shih-Fu Chang, "Image Retrieval: Past, Present and Future", Journal of Visual Communication and Image Representation, Vol. 10, pp. 1 - 23, 1999.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....in all our experiments. 1. INTRODUCTION Content based information retrieval (CBIR) has attracted a lot of research interest in recent years. A typical CBIR system, e.g. an image retrieval system, includes three major aspects: feature extraction, high dimensional indexing, and system design [1]. Among the three aspects, feature extraction is the basis of content based information retrieval. However, features we can extract from the data are often low level features. We call them low level features because most of them are extracted directly from digital representations of objects in the ....

Yong Rui, Thomas S. Huang, and Shih-Fu Chang, "Image Retrieval: Past, Present, and Future", Proceeding of International Symposium on Multimedia Information Processing, Dec. 1997.

....model retrieval. I. INTRODUCTION C ONTENT BASED INFORMATION RETRIEVAL (CBIR) has attracted a lot of research interest in recent years. A typical CBIR system, e.g. an image retrieval system, includes three major aspects: feature extraction, high dimensional indexing, and system design [1]. Among the three aspects, feature extraction is the basis of CBIR. However, features we can extract from the data are often low level features. As a result, two semantically similar objects may lie far from each other in the feature space, while two completely different objects may stay close to ....

Y. Rui, T. S. Huang, and S.-F. Chang, "Image retrieval: Past, present, and future," in Proc. ISMIP, Dec. 1997.

....are IBM s QBIC [3] Virage [1] VisualSEEk [12] and Photobook [7] All these systems ask the user to select feature(s) and sometimes, suitable weights. They offer Query by Example and (most of them) sketch based queries. Almost all CBIR systems follow the so called computer centric approach (CCA, [8]) querying is done with feature and distance functions where the user has to select features for a query and the weights to determine their relative importance. The CCA has two major drawbacks: The semantic gap This term addresses the difference between high level concepts for CBIR and the ....

Rui, Y., Huang,T., Chang, S., "Image Retrieval: Past, Present and Future", International Symposium on Multimedia Information Processing, Taiwan, 1997.

.... Key words: Multimedia Databases, Processing a Multimedia Joins, Nearest Neighbor Search 1 Introduction and Motivation Commonly used content retrieval systems focus on the problem of finding the nearest neighbor (NN search) for a given single query object out of a database of media objects [1]. However, there are only few attempts [2,3] that consider join operations on two multimedia tables, where the multimedia data components are represented by their respective feature vectors. The necessity of using multimedia joins in a variety of applications is the motivation behind this search ....

....join with RES is performed. 2 Problem Definition In an image database, a k Nearest Neighbor (k NN) search method is the retrieval of a set of k (k # 1) objects from a larger set of image data objects that are similar to a given query object, using their feature vector representations [1]. Formally, the k NN search is defined as follows : Definition 1 (k Nearest Neighbor Search) Given a set of images S, a query image q, and a positive integer k; the k nearest neighbors to the query image q denoted as NN k (S, q) are the first k images that are a shorter distance from q in the ....

[Article contains additional citation context not shown here]

Y. Rui, T. S. Huang, and S.-F. Chang. Image retrieval: Past present and future. Journal of Visual Communication and Image Representation, 10:1--23, 1999.

....expected that frames that cluster well with each other will come from the same scene. Pixel based clustering is a crude method and feature based methods for this purpose are to be preferred. Clustering can be performed using a range of video features including region shape, colour and texture [7]. One method of clustering frames can be based on the use of global texture. Here we are interested in gray scale texture (for colour texture features, see [4] As successive frames in the same scene change only slightly in their features such as their global texture, we expect that if we are to ....

Y. Rui, T.S. Huang and S.F. Chang, "Image retrieval: past, present and future", Journal of Visual Communication and Image Representation, vol. 10, pp. 1-23, 1999.

....Database, April 2001. I. Introduction Content based information retrieval (CBIR) has attracted a lot of research interest in recent years. A typical CBIR system, e.g. an image retrieval system, includes three major aspects: feature extraction, high dimensional indexing, and system design [1]. Among the three aspects, feature extraction is the basis of content based information retrieval. However, features we can extract from the data are often low level features. We call these low level features because most of them are extracted directly from digital representations of objects in ....

Yong Rui, Thomas S. Huang, and Shih-Fu Chang, "Image Retrieval: Past, Present, and Future", Proceeding of International Symposium on Multimedia Information Processing, Dec. 1997.

....which requires 500 disk accesses, but 90 of the images retrieved are relevant. Thus in order to evaluate the relative performance of two different algorithms, we need a measure combining the retrieval performance and execution cost. Although, it is recognized as an important problem as in [28], to the best of our knowledge, none of the CBIR literature has addressed this issue. Smith has used execution cost measure to evaluate the performance of retrieval system to find the right images [32] Ortega has evaluated the retrieval performance using precision and recall [24] The question ....

Yong Rui, Thomas S. Huang and Shih-Fu Chang. Image retrieval: Past present and future. Journal of Visual Communication and Image Representation, Volume 10, pages 1 -- 23, 1999.

....engines, such as Yahoo, Alta Vista, Lycos, etc. At the early stage of CBIR, research primarily focused on exploring various feature representations, hoping to find a best representation for each feature. For example, for texture feature alone, almost a dozen representations have been proposed [21], including Tamura [22] MSAR [23] Word decomposition [24] Fractal [25] Gabor Filter [26, 11] and Wavelets [27, 28, 12] etc. The corresponding system design strategy for early CBIR systems is to first find the best representations for the visual features. Then, ffl During the retrieval ....

....knowledge of the low level feature representations used in the retrieval system, which is normally not the case. Motivated by the limitations of the computer centric approach, recently research focus in CBIR has moved to interactive mechanism that involves human as part of the retrieval process [21, 29, 30]. Examples include interactive region segmentation [31] interactive image annotation [29, 32] usage of supervised learning before the retrieval [33, 34] and interactive integration of keywords and high level concepts to enhance image retrieval performance [10, 35] In this paper, to overcome ....

T. S. Huang and Y. Rui, "Image retrieval: Past, present, and future," in Proc. of Int. Symposium on Multimedia Information Processing, Dec 1997.

No context found.

Yong Rui, Thomas S. Huang and Shih-Fu Chang, "Image Retrieval: Past, Present and Future", Journal of Visual Communication and Image Representation, Vol. 10, pp. 1 - 23, 1999.

No context found.

Y. Rui, T. S. Huang, and S.-F. Chang. Image retrieval: Past present and future. Journal of Visual Communication and Image Representation, 10:1--23, 1999.

No context found.

Y. Rui, T. S. Huang, and S.-F. Chang. Image retrieval: Past present and future. Journal of Visual Communication and Image Representation, 10:1--23, 1999.

No context found.

Y. Rui, T.S. Huang, S.F. Chang, "Image Retrieval: Past, Present, And Future", Proceedings International Symposium on Multimedia Information Processing, 1997.

No context found.

Y. Rui, T. S. Huang, S.-F.Chang, "Image Retrieval: Past, Present, and Future", Journal of Visual Communication and Image Representation, 10:1-23, 1999.

No context found.

Y. Rui, T. S. Huang, and S.-F. Chang, "Image retrieval: Past, present, and future," J. Visual Commun. Image Represent. 10,39--62#1998#.

No context found.

Y. Rui, T.S. Huang and S-F. Chang, "Image Retrieval: Past, Present, and Future," Journal of Visual Communication and Image Representation, Vol. 10, pages 1-23, 1999.

No context found.

Rui, Y., Huang, T. S., Chang, S.-F.: Image retrieval: Past present and future. J. of Visual Communication and Image Representation, Vol. 10. (1999) 1-23

No context found.

Yong Rui, Thomas S. Huang, and Shih-Fu Chang, "Image Retrieval: Past, Present, and Future", Proceeding of International Symposium on Multimedia Information Processing, Dec. 1997.

No context found.

Y. Rui, T.S. Huang, S.F. Chang, "Image Retrieval: Past, Present, And Future", Proceedings International Symposium on Multimedia Information Processing, 1997.

No context found.

Y. Rui, T. S. Huang, S.-F.Chang, "Image Retrieval: Past, Present, and Future", Journal of Visual Communication and Image Representation, 10:1-23, 1999.

No context found.

Yong Rui, Thomas S. Huang, and Shih-Fu Chang, "Image Retrieval: Past, Present, and Future", Proceeding of International Symposium on Multimedia Information Processing, Dec. 1997.

No context found.

Y. Rui, T. S. Huang, and S.-F. Chang, "Image Retrieval: Past, Present, and Future",

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