ABSTRACT
Several types of moving images are available now, and each has its own indexing needs. In addition, a number of levels of indexing are necessary, depending on the type of image, the type of collection, and the needs of users. In information science, work on providing indexing to the various levels has largely to do with finding ways to recycle text created for other purposes in the processes of production, in order to point to individual shots, sequences, scenes, or chapters. Such text recycling needs to happen automatically, through the application of algorithms developed for this purpose, since indexing at the various levels by humans is prohibitively expensive in most circumstances. Multilingual indexing is an issue in the context of retrieving images in a networked environment. Another issue is access to moving images using indexing approaches other than subject indexing. Tagging of images by users is prevalent in the networked environment, and a discussion of its usefulness is included. Finally, there is some speculation on what the future of moving image indexing might bring.
