IMPART will research, develop and evaluate intelligent information management solutions for ‘big data’ problems in the field of digital cinema production. It will find new ways of managing, visualising and analysing very large multimodal data sets so that creative personnel can review three-dimensional scene representations on the set, understand the data, identify errors, evaluate the quality of the shot and take creative decisions in real time. We will research and test solutions for the use case of large-scale movie production, with very large research and evaluation datasets based on the latest industrial processes. Results will be applicable to film production and more generally to commercial and professional domains ranging from TV and on-line video to surveillance. The project outcomes will include: • Architectures and file structures for intelligent multimodal data and metadata management, presentation and browsing linking data, metadata and advanced applications • Methods for automatically selecting relevant subsets of data to allow rapid processing • Techniques for processing, information extraction and analysis in distributed systems • Multimodal content semantic and low-level analysis techniques for the automatic derivation of metadata to support visualisation, decision support, management and retrieval • On-set decision support, including tools for multimodal data visualisation, data summarization, layout, registration and cross-referencing, data and metadata search, virtual overlays and real-time previews of effects, creative and technical decision support • A demonstrator system for professional testing and evaluation, both on and off set • A documented xml interface that allows third parties to develop compatible tools and processes • A published test dataset of multimodal data and metadata captured in realistic film production conditions for use by other researchers.