MediaMap will act on 3 tied services in AV production:
In order to create these services, 2 main problems have to be solved: creating an efficient collaborative workflow and raising the quality level of amateurs’ production as close as possible to the professional level.
Under the current workflow model, an audiovisual project cannot be considered as a whole, and the different steps in the process cannot be considered and used separately. The audiovisual chain is sequential, and constituted of different steps: scripting, shooting, digitising, editing, compositing, subtitling, mixing, broadcasting and, finally, indexing before publishing on the Web.
It is only at the end that everybody has a complete view of the project.
A collaborative process has to be different. There is a need to allow different users to access different parts of the ongoing work at a different time and to share both the content and the tasks to be performed. There is a need for any contributor to be able to know the editorial intention at each stage of the work.
Such collaboration will be possible only if everyone can access the asset as it is, in progress, with all the editorial intentions at each step, even before filming. Such collaboration will be possible only if indexation occurs early in the process, at the very beginning of the chain, before filming.
The first part of the solution proposed in MediaMap consists in creating a new wrapper for AV production, whose elements will be accessible by all collaborators. The wrapper will include not only video, audio and metadata elements, but also a description of the structure of the movie, the editorial intentions, targeted metadata, the realization method, etc. It could be seen as a “video template” fully defined at the beginning and then simply filled by all collaborators.
The second part of the solution is to give access to this new container to all collaborators through a “documentary bus” relying on a network’s infrastructure upon which its elements will easily circulate.
Filling this video template is simplifying dramatically the usual workflow and gives more efficiency to operators that know before filming what and how they have to shoot.
In a few words, it brings the (hyper)text editing method to (hyper)video production. However, where text input requires a mouse and a keyboard, video production needs a new type of interactive tool: an “ingest” function for capturing audio and video supported by a PC or by new type of camcorder which is not only a shooting device, but also a complete network terminal, guiding, helping or correcting the operator in real time when filling the “video template”.
The “documentary bus” concept is borrowed from the OAIS (Open Archival Information System) ISO standard, which deals with the functional specifications of a production / broadcast model that allows retrieval, transaction, exchange, etc. of a self-described content.
The documentary bus implements the technical specifications of AXIS (as defined in the Memories European project), where every element of the process is open and autonomous (concept of Autonomous Asset Entities, AAE).
MediaMap will add technical information and semantic content description to these (AXIS) AAE. The documentary bus will be changed into a “Open Semantic Bus”, and Autonomous Asset entities in Unique Semantic Entities or USEs, which will allow giving access to content and sharing editorial intentions throughout a project. Even based on this complex structure, a USE will be seen simply as “the template” described above by the operator.
Three tools will be developed to perform the operations: a software ingest tool allowing collaboration from a PC station, an interactive networking camcorder (called hereafter Camcorder 2.0) allowing standalone nomadic production and a piece of hardware and software (called hereafter the Camera Mate) that will offer the same functionality as the Camcorder 2.0 but will be connected to a standard camcorder. The semantic camcorder 2.0 will include a very innovative Man Machine Interface to guide and totally assist the operator as predefined in the “template”, including example display, text capture, etc. But the camcorder will also include specific algorithms taking care automatically of all parameters to make good video and audio capture, and never miss it. It will also support some effects like zooming.
The properties of the USE will allow MediaMap to care about the persistence of content, legal rights and multilingual issues.
Finally, MediaMap will propose new and efficient publishing possibilities; it will allow a new way of accessing contents through index and metadata that are really adapted to the viewers’ profiles. It will include a new topographical approach for browsing within contents themselves and within different “territories” thanks to a semantic search engine.
A side effect of such a semantic approach will be to provide a better way of financing content web sites based on it by allowing the power of a brand new advertisement model, that would be highly fine-tuned to both the content and the user.
MediaMap Project is, first of all, a “Practice Project”. It will allow dealing with content settings and publishing intentions quite early in a project, even at the conception stage, in order to pilot and structure the whole development of the project.
It will bring adapted and innovative solutions: