Media Computing

With the explosion of digital media and online services, a key challenge is the automation of content annotation, indexing and organizing for efficient media access, retrieval and browsing. A major failing of automated media annotation systems is the semantic gap - the incompatibility between the low-level features that can be computed to describe media and the high-level meaning sought by user queries in searches for audio and video. This text explores the area of computational media aesthetics to bridge the semantic gap. It brings together researchers from multiple disciplines, from computer scientists to content creators from engineers to media specialists, to highlight advances in this area. It examines the elements of media production and aesthetics principles in order to establish the foundation for semantics indexing, and to build innovative and effective technologies for content annotation, search and browsing. The book outlines strategic means and methods to derive high level semantic constructs from automatic analysis of media. It is suitable for the use as a text in courses on multimedia and informational retrieval.