State of The Art: Informedia Digital Video Library

Abstract

This chapter provides a description of the most comprehensive searchable digital video library in the state of the art, the Informedia project at CMU Informedia,; Wactlar et al., 1996b. This project is comprehensive in terms of the amount of digital video that it has archived in various content searchable representations — Informedia currently has about 1,000 hours (1 terabyte) of video that has been automatically processed by integrated speech, image, and language understanding technologies. In its efforts to create an automatically processed and large-scale video collection, Informedia is not as ambitious, in terms of the content analysis, as the technologies “on the horizon” described in the previous chapter. In terms of the CBAM technical components (Figure 2.6), Informedia represents the state of the art for content representations offering semantic analysis of the content in large video repositories. It is a testament to the immaturity of video content management that Informedia can only offer fairly shallow exploitation of the content in video repositories, as will be described in this chapter.