Summary and recommendations

Abstract

There is a host of open issues in the graphics recognition research area, and there is no doubt as to the viability and growing importance of the field. Contribution has already been made in specialized areas, for instance, for telephone and power companies which hold huge numbers of drawings with the same syntax and appearance, making the development of a specialized system for processing them cost effective. Image and graphical document databases are two multimedia-enabling technologies that seem to be heavy consumers of graphic based operations such as graphical queries. Likewise, the bi-directionality of exchanging information stored on paper and electronic media will make increasing use of graphic recognition capabilities. Overall, a technology that was originally motivated by the pressing need to convert paper documents containing graphics into electronic formats is becoming more and more useful in a variety of information technology domains never before thought of. Once the technology becomes mature enough, one can envision a team of humans working with robots, where both humans and robots read the same graphical document (displayed on any media) and coordinate their tasks based on understanding the same document.