J. R. Bach, S. Paul, R. Jain, A Visual Information Management System for the Interactive Retrieval of Faces, IEEE Transactions on Knowledge and Data Engineering, 5(4), Aug. 1993, pp. 619-628.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....visual features such as color, texture, shape, and or motion to search large databases (see [1, 2, 3] for recent surveys on image and video indexing and retrieval technologies) These systems may be classified as special and general purpose systems. Special purpose systems include: Xenomania [4] for face image retrieval based on query by example; Trademark image database system [5] for trademark retrieval using shape information; and the Center of Excellence for Document Analysis and Recognition (CEDAR) system [6] for indexing and retrieving documents by understanding captions. General ....

J. R. Bach, S. Paul, and R. Jain, "A visual information management system for the interactive retrieval of faces", IEEE Trans. on Know. and Data Eng., vol. 5, no. 4, Aug. 1993.

....may specify a fixed number of labeled objects and a variable number of unlabeled objects. Notice that this is a general setting. One special case is the case where all objects are unlabeled (k = 0) Another special case is when all objects are labeled (u = 0) in all images as was the case in [2]. There, the problem was to search a database of face images; from each image, a fixed number of labeled objects are identified (eyes, nose, etc. and their attributes and relative positions are computed. 2.1 Background Image descriptions are given in terms of object properties and in terms of ....

....of the features and the matching function, with little or no emphasis on the speed of retrieval. Thus, each image is described by a set of features; to respond to a query, the system searches the features of all the images sequentially. A typical, recent system along these lines is described in [2]. The system supports the segmentation and interactive retrieval of facial images from an IDB. A priori knowledge regarding the kind and the positioning of expected image objects (e.g. face outline, nose, eyes etc. is employed and used to guide the segmentation of face images into disjoint ....

Jeffrey R. Bach, Santanu Paul, and Ramesh Jain. A Visual Information Management System for the Interactive Retrieval of Faces. IEEE Trans. on Knowledge and Data Engineering, 5(4):619--627, August 1993.

....has known a significant rise for several reasons: Graphical properties are automatically calculated, Graphical interrogation consists of comparing two images vectors, Sorting of result by ordering images according to their relevance. Systems adopting this approach are numerous such as [Jef 93] Ash 95] Cio 98] Gel 98] Che 98] Hue 98] and [Sch 99] In spite of the great current interests of the graphical approach, it is still too open ended to be appropriate in several domains for the following reasons mentioned underneath: 1. Searching for shape, like circle for example, in ....

Jeffrey R. Bach, Santanu P., Ramesh J., "A Visual Information Management System for the Interactive Retrieval of Faces", IEEE transactions on Knowledge and data engineering, Vol. 5, N 4, August 1993.

....domain knowledge appearing in form of classes, that manages image features and operators semantics during query interpretation. Other approaches are based on fuzzy searching, taking into account the subjective interpretation of image features [11] and domain specific image distinctive landmarks [12]. In general, databases and retrieval systems designed for specific application fields can use domain knowledge in several forms, in order to improve the classification and retrieval processes. In [13, 14] segmenting techniques of video clips are based on content analysis for identifying the ....

J. Bach, S. Paul, R. Jain. (1993). A Visual Information Management System for The Interactive Retrieval of Faces. IEEE Trans. on Knowledge and Data Engineering, 5, 4, 619-628,

.... acquisition, storage, indexing and retrieval of map images [Samet, So er 1996] real time ngerprint matching from a very large database [Ratha et al. 1996] querying and retrieval using partially decoded JPEG data and keys [Schneier, Abdel Mottaleb 1996] and retrieval of faces from a database [Bach et al. 1993], Wu, Narasimhalu 1994] Finding documents on the Web which have images of interest is a much more sophisticated problem. Two well known portals which have a search interface for a database of images are: Yahoo Image Surfer 105 and Alta Vista PhotoFinder 106 . Like Yahoo s text based ....

Bach, J., Paul, S., Jain, R., \A visual information management system for the interactive retrieval of faces" IEEE Trans. Knowledge and Data Engineering, 5, 4 (Aug. 1993), 619-628.

....difference, and (b) how to represent a single shape compactly. We address (a) in Section 3.3. With respect to (b) the most popular methods are: ffl representation through landmarks : for example, in order to match two faces, information about the eyes, nose, etc. are extracted manually [4] or automatically. Thus, a shape is represented by a set of landmarks and their attributes (area, perimeter, relative position, etc) The distance between two images is the sum of the penalties for the differences of the landmarks. ffl representation through numerical vectors, such as (a) samples ....

Jeffrey R. Bach, Santanu Paul, and Ramesh Jain. A visual information management system for the interactive retrieval of faces. IEEE Trans. on Knowledge and Data Engineering (TKDE), 5(4):619--628, August 1993.

.... direct impact on the progress of the revolution in communication precipitated with the increasing availability of digital video [7] The complexity in the very nature of two dimensional image data gives rise to a host of problems that alphanumeric information systems were never designed to handle [1]. A central task of these multimedia information systems is the storage, retrieval, and management of images [10] In many cases, the operator would like to base this retrieval on objects contained in the images of the database. As such, content based image retrieval is an object recognition ....

J. R. Bach, S. Paul, and R. Jain, "A visual information management system for the interactive retrieval of faces," IEEE Transactions of Knowledge and Data Engineering, vol. 5, no. 4, p. 619ff, 1993.

....part of VIMSYS system. Support for modeling image sequences is provided. Sequences can be based on spatial contiguity of images or images of the same geographic region recorded at dierent time points. An application of VIMSYS data model for interactive retrieval of face information is described in [10]. REMINDS is proposed as a generic image database management system that includes an image interpretation module [122] Though the underlying data model is relational, REMINDS has several structural object orientation features. Model base module is at the core of the image interpretation task and ....

....extremely impractical in image database systems due to the following. In industrial applications, objects to be recognized are limited to few domain objects. Either models (in the case of model driven approach) or feature trees (in the case of data driven approach) are used to assist in this task [68, 67, 64, 10]. In several image database applications, since the objects to be recognized originate from disparate domains and images contain considerable noise, model based or data driven approaches to object recognition may not be highly useful. On the other hand, a completely manual approach to object ....

J. Bach, S. Paul, and R. Jain. A visual information management system for interactive retrieval of faces. IEEE Transactions on Knowledge and Data Engineering, 5(4):619-628, 1993.

....and implementation of image database systems supporting queries by image content are image feature extraction[7, 10, 13, 15, 17] image content representation and organisation of stored information[42, 43] CHAPTER 2. PROBLEM FORMULATION AND CONTRIBUTIONS 10 searching and retrieving strategies[6, 25, 40, 41], and user interface design. Addressing such issues has become the object of intensive research activities in many areas of computer science over the past few years. Advances mainly in the areas of databases and computer vision research have resulted in methods which can be used for image ....

....Methods that use linear quad trees [4] ffl Methods that are based on trees. One of the most characteristic approaches is the R tree [5] A system designed to support the segmentation, the description, as well as the interactive retrieval of facial images from an image database is presented in [6]. A priori knowledge regarding the kind and the positioning of expected image objects is employed and used to guide the segmentation of face images into disjoint regions corresponding to the above objects. Retrieval is performed in stages allowing the user to adjust the query criteria at each ....

J. Bach, S. Paul, and R. Jain, "A visual information management system for the interactive retrieval of faces", IEEE Trans. on knowledge and data engineering, V.5, #4 pp. 619-627, 1993. 112 BIBLIOGRAPHY 113

.... which are is located within a certain distance from a type B symbol; real time ngerprint matching from a very large database [Ratha et al. 1996] querying and retrieval using partially decoded JPEG data and keys [Schneier Abdel Mottaleb 1996] and retrieval of faces from a database [Bach et al. 1993], Wu, Narasimhalu 1994] 7.4.2 Still Images in a Video Scene A research area closely related to retrieval of still images from a very large image database is query and retrieval of images in a video frame or frames. We mention just a few research projects in this area below. Some examples ....

Bach, J., Paul, S., Jain, R., \A visual information management system for the interactive retrieval of faces", IEEE Trans. Knowledge and Data Engineering, 5, 4 (Aug. 1993), 619-628.

....with conceptual terms and predicates. Thus, we use a decomposition approach to describe the specific shape features of interested objects. Statistical approaches can be used to retrieve similar images by relaxing a certain percentage of the standard deviation of the feature values (e.g. VIMS [BPJ93] However, the same amount of relaxation is applied throughout the distribution, and thus is insensitive to the position of the operating point. Moreover, many image features are based on multiple attributes. Using standard deviation to retrieve similar feature values lacks the consideration of ....

J. R. Bach, S. Paul, and R. Jain. A visual information management system for the interactive retrieval of faces. IEEE Transaction on Knowledge and Data Engineering, Oct 1993.

....domain of images. The type of information in the summary is determined at the time the system is built and can t be changed. Images in a given specialized domain have much in common. This commonality is reflected in the type of information put in summaries. For example, the VIMS (or VIMSYS) [1, 3] system can handle only images of frontal views of single Human faces. The type of information in the summaries fits this image domain. A special feature extraction algorithm was constructed so that summaries can be created automatically. The query language is designed for this specific image ....

Jefferey R. Bach, Santanu Paul and Ramesh Jain. A visual information management system for the interactive retrieval of faces. IEEE Transactions on Knowledge and Data Engineering, 5(4):619-- 628, August 1993.

....In these databases, categorization of data is straightforward on the basis of their semantics. Efficient data access can be facilitated by B trees and other data structures. However, such traditional approaches to indexing may not be appropriate in the context of content based image retrieval [1, 4, 18, 12, 5, 7, 2, 19]. In this context, a challenging problem arises with many image databases, within which queries are posed via visual or pictorial examples (termed visual queries) A typical visual query might entail the location of all images in a database that contain a subimage similar to a given query image. ....

....to support content based image retrieval. Thus, an image database can be assumed to consist of feature vectors. Substantial research has been directed toward the support of efficient indexing techniques based on feature This research is supported by Xerox Corporation. vectors of images [2, 14, 17, 15, 3, 9]. Given feature vectors of images, indexing algorithms such as R tree, R tree, and TV tree [15, 3, 9] have been proposed to support efficient accesses to image databases. In a large scale image database, a critical problem to be solved is to classify feature vectors into different ....

J.R. Bach, S. Paul, and R. Jain. A Visual Information Management System for the Interactive Retrieval of Faces. IEEE Transactions on Knowledge and Data Engineering, 5(4):619--628, 1993.

....in the color of the sky. In the latter case the images themselves are not restricted, but the isomorphism between images and meanings is. An unrestricted database, of course, is composed of images that are not restricted neither in content nor in interpretation. For instance, 38] 37] 29] and [2] can be considered restricted in our sense, while [23] 12] and [15] are unrestricted. There is no place in the image to which we could ascribe the meaning or parts of it (from which particular part of a tree does its shape come from ) In an image of the sea at twilight, there is no pixel ....

....3: Graphs of the distance function obtained from the Tversky similarity. The figures refer to a two dimensional measurement space and two dimensional predicate space. The matrix A is the 2 2 identity matrix. The four figures are distances with respect to the references r = 0, 0] T , r = [0, 2] T , r = 1, 1] T , and r = 1, 4.5] T , respectively. 16 Figure 4: Length of the tangent vector [1 2, 1 2] in the perceptual space The figures refer to a twodimensional measurement space and two dimensional predicate space. The matrix A is the 22 identity matrix. The four figures are ....

[Article contains additional citation context not shown here]

J.R. Bach, S. Paul, and R Jain. A visual information management system for the interactive retrieval of faces. IEEE Transactions on Knowledge and Data Engineering, 5(4):619--628, 1993.

No context found.

J. R. Bach, S. Paul, R. Jain, A Visual Information Management System for the Interactive Retrieval of Faces, IEEE Transactions on Knowledge and Data Engineering, 5(4), Aug. 1993, pp. 619-628.

No context found.

J. R. Bach, S. Paul, R. Jain, A Visual Information Management System for the Interactive Retrieval of Faces, IEEE Transactions on Knowledge and Data Engineering, 5(4), Aug. 1993, pp. 619-628.

No context found.

J. Bach, S. Paul, R. Jain, "A Visual Information Management System for The Interactive Retrieval of Faces", IEEE Trans. on Knowledge and Data Engineering, 5, 4, 1993

No context found.

J. Bach, S. Paul and R. Jain, A Visual Information Management System for The Interactive Retrieval of Faces, IEEE Trans. on Knowledge and Data Engineering, 5, 4, 1993

No context found.

J. R. Bach, S. Paul, and R. Jain. A visual information management system for the interactive retrieval of faces. IEEE Transaction on Knowledge and Data Engineering, August 1993.

No context found.

J. Bach, S. Paul, R. Jain, "A Visual Information Management System for the Interactive Retrieval of Faces", IEEE Trans. on Knowledge and Data Engineering, Vol.5, No.4, Aug.1993.

No context found.

J. Bach, S. Paul, R. Jain, A Visual Information Management System for The Interactive Retrieval of Faces, IEEE Trans. on Knowledge and Data Engineering, 5, 4, 619-628, 1993

No context found.

Bach,J.R., Paul,S., Jain,R., \A Visual Information Management System For The Interactive Retrieval Of Faces", IEEE Transactions on Knowledge and Data Engineering, vol. 5, no. 4, pp. 619-628, August 1993.

No context found.

J. R. Vach, S. Paul and R. Jain, A Visual Information Management System for the Interactive Retrieval of Faces, IEEE Transactions, Knowledge and Data Engineering, Oct 1993.

No context found.

J. R. Bach, S. Paul, and R. Jain, "A visual information management system for the interactive retrieval of faces," IEEE Trans. on Knowledge and Data Engineering, vol. 5, no. 4, pp. 619--628, Aug. 1993.

No context found.

J. Bach, S. Paul, R. Jain, "A Visual Information Management System for the Interactive Retrieval of Faces", IEEE Trans. on Knowledge and Data Engineering, Vol.5, No.4, Aug.1993.

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