Enhancing usability of raw media materials
RUSHES: Retrieval of Multimedia Semantic Units for Enhanced Reusability
The overall aim of RUSHES is to design, implement, validate, and trial a system for both the delivery of and access to raw media material (rushes), as well as its reuse in the production of new multimedia assets. RUSHES targets indexing, search and retrieval of rushes archives, to ease in-house postproduction and reuse in a professional environment. It also aims at providing raw content to end users for the creation and editing of content in a shared computational environment using a multimodal approach that combines knowledge extracted from the original content, metadata, visualization and interactive browsing functionality.
RUSHES will consider professional content creator or provider, and consumer for the design of the system. At the provider side, techniques for automatic content cataloguing and semantic based indexing will be used to link raw content with metadata. At the consumer side, browsing and retrieval strategies will be implemented to provide advanced search functionalities and low access latency when navigating rushes databases. The core research issues of RUSHES are knowledge extraction for semantic inference, semantic-based content annotation, scalable multimedia cataloguing, interactive navigation, and non-linear querying and retrieval techniques using hierarchic descriptors.
To evidence the viability of the project goals, a complete system for search, retrieval and use of rushes is designed, implemented and tested with potential professional and end-users. The envisaged system prototype does not only serve as a means to showcase the functionality of the proposed technology but also provides a realistic and commercially valid proof of concept.