A. Abella and J. Kender. From pictures to words: Generating locative descriptions of objects in an image. ARPA94, pages II:909--918, 1994.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....or text based image retrieval systems rely on manual annotations, captions, document context, or pre generated keywords, which leads to a loss of exibility through the initial choice of annotation and indexing. Languages mainly concerned with deriving textual descriptions of image content ([1]) are inappropriate for general purpose retrieval since it is infeasible to generate exhaustive textual representations which contain all the information and levels of detail which might be required to process a given query. Formal query languages such as extensions of SQL [9] are limited in their ....

A. Abella and J.R. Kender. From pictures to words: Generating locative descriptions of objects in an image. In ARPA94, pages II:909-918, 1994.

....up with my own analysis. The version of surge resulting from these extensions has since been used for other generation applications in addition to streak: automated documentation for the activity of telephone network planning engineers [ Kukich et al. 1994 ] verbal descriptions of visual scenes [ Abella 1994 ] and generation of business letters. The two sets of extensions to the fuf surge package are described in detail Appendix B. They were essentially preparatory work paving the way for development of streak as a prototype for a particular application. The present chapter describes this development ....

....up with my own analysis. The version of surge resulting from these extensions has since been used for other generation applications in addition to streak: automated documentation for the activity of telephone network planning engineers [ Kukich et al. 1994 ] verbal descriptions of visual scenes [ Abella 1994 ] and generation of business letters. In what follows, I first give an overview of fuf (Functional Unification Formalism) as a programming language, including the extensions for non monotonic processing prompted by the development of streak. I then describe the initial version of surge (Systemic ....

[Article contains additional citation context not shown here]

A. Abella. From pictures to words: generating locative descriptions of objects in an image. In Proceedings of the Image Understanding Workshop, Monterrey, CA, November 1994.

....detail the implementation of the revision rules. 22 Since its development, surge 2. 0 has been used as front end in other generation applications than streak, including automated documentation for the activity of telephone network planning engineers [28] and verbal descriptions of visual scenes [1]. 23 These words include neither proper nouns and quantitative values which are passed from the input to the generator, nor function words which get added by surge (surge covers about 100 function words) 2.4.2 Encoding and interpreting revision rules Revision rules are represented in an FG ....

A. Abella. From pictures to words: generating locative descriptions of objects in an image. In Proceedings of the Image Understanding Workshop, Monterey, CA, November 1994.

....Blocks World [Winograd, 72] where arrangements of blocks were described. Srihari s PICTION system [Srihari, 1991] extracts faces from photos by parsing newspaper captions, and [Neumann, 1984] and [Marburger et al., 1981] extract natural language descriptions of traffic scenes. More recently, [Abella and Kender, 1994] report a system that describes relative locations of objects. All of these systems operate in specialized domains and use limited vocabularies. V. Summary Assessment and Future Research Overall the Image Context Vector approach looks promising. Its strong points are: sensitivity to embedded ....

Abella A and Kender JR (1994). From pictures to words: generating locative descriptions of objects in an image. Image Understanding Workshop, Monterey, CA. Morgan Kaufmann, 909-918.

....that locate the stone and register the X ray with a model of the X ray are described, and how the language generation preprocessor adheres to the appropriate medical terminology is demonstrated. Several examples illustrate the results of the image processing and language generation preprocessor [3]. The goal is to describe the location of renal stones found in radiographs, by observing their spatial relationship to other parts of the radiograph. Those parts include the spinal cord, the various lumbar bodies of the spinal cord, the bladder, the kidney, the calyx, and a few others that are ....

Alicia Abella and John R. Kender. From pictures to words: Generating locative descriptions of objects in an image. In Image Understanding Workshop, 1994.

....the X ray with a model of the X ray. We will also demonstrate how the language generation preprocessor adheres to the appropriate medical terminology when generating descriptions. Several examples will be given to illustrate the results of the image processing and language generation preprocessor [Abella and Kender, 1994b] 7.1 Introduction The goal of this task is to describe renal stones found in radiographs the precise goal of a radiologist examining an X ray. The issue in the exam is to first determine if a stone is present in the radiograph, and if so to identify its location for possible treatment. This ....

Alicia Abella and John R. Kender. From pictures to words: Generating locative descriptions of objects in an image. In Image Understanding Workshop, 1994.

No context found.

A. Abella and J. Kender. From pictures to words: Generating locative descriptions of objects in an image. ARPA94, pages II:909--918, 1994.

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