In Frame Time Based Descriptive (IFTBD) Metadata

This feature article was contributed by Peter Flood, CEO, AIDA Content Management, Inc


I recently viewed an excellent Interactive Live Q&A Video Conference: State of DAM Today. The Panelists were David Diamond, Carin Forman, Ralph Windsor and the Moderator was Frank DeCarlo. The conference discussed the current but persistent challenges existing in the DAM world today – Promises of DAM Vendors; AI/Machine Learning promises; Trust in Metadata whether manual or automated; Its application in other worlds such as CRM, Martech, Ecommerce; The human aspect of process; Level of Search enabled. And I certainly in agreement with all.

The key tent pole in the discussion though centered on the current state of innovation in DAM and as one panelist stated, ‘There’s been virtually no innovation and nothing interesting happening in DAM since really the beginning.’ And I agreed with the perception but not the reality.

There is a solution innovation which became commercial over a year and half ago. In fact, this innovation is substantive enough to argue it is a whole new disruptive category to itself. This is In Frame Time Based Descriptive (IFTBD) metadata – a term that describes the key capability of the VRmeta solution suite. It was created by Mark Milstein, Managing Director of Microstocksolutions,

The purpose of this article is to ‘articulate the science’ as it relates to the application of IFTBD metadata and its value to content management and distribution which spans all industries that utilize digital assets. Its application, integration and implementation within the content management area, albeit smooth and fluid, will be the subject of another article.

In Frame Time Based Descriptive Metadata (IFTBD) is the capability to input time-based descriptive metadata to any person, object or event, and record that meta-tag as either an in-frame metric and/or based upon the asset time code. This allows for an assets multiple distinct definable characteristic to be placed within the asset as an in-frame metric.

The descriptive capacity allows for every definable data element including Administrative, Technical, Rights Management, Preservation/Archive, Discovery metadata with any frame point. Thus, complete metadata ‘ground truth’ of a digital asset can be captured for the record during the assets life cycle and more importantly optimizes search, discovery and marketability to a level that has not existed before.

This level of optimization is further enhanced given the description length capability for any tag is near unlimited and output data include object-based tag orientation in frame with in and out times and duration. Further creating actionable metadata through the asset’s lifecycle for multi-use.

The underlying capability allows for adding a 2000-character long caption or summary into a universally legible Dublin Core Metadata Initiative/IPTC™ core description field. Both Dublin Core and IPTC™ metadata standards enjoy universal acceptance. The two schemas define metadata structure, properties, and fields, so that whatever media it’s applied to is optimally described and easily accessed later.

The toolset in which this capability resides allows for the insertion of metadata from the point of creation of an asset through all workflows to the point where it might be retired or archived. In addition, the toolset allows for the application of metadata as simply time based (frame level) or as Captions and IPTC keywords. All metadata including IFTBD can be outputted as CSV File, Embed XMP or XMP sidecar and move smoothly through all workflows and into MAM/DAM.

The IFTBD capacity creates its own organizational ability for optimized search particularly if controlled vocabularies and taxonomies are employed. It circumvents the common process of applying metadata within the digital asset management system with limiting field constraints as the only method to enter metadata.

IFTBD metadata can be applied to all digital asset formats including still photographs (not time based), linear video and immersive/360 content. This is important to immersive/360 content given the area covered in a frame is 5x larger than that covered in a linear or standard video frame.

Finally, AI and Machine Learning gain significant benefits from IFTBD metadata. In this current educational phase of training Machine Learning models, relevance of output is an enormous challenge and will persist until Machine Learning models are smart enough to allow a transition to the value-add predictive phase which is the goal. The ability to apply IFTBD metadata to a digital asset which is then fed into a Machine Learning training flow deepening the AI’s capability on an accelerated basis..

IFTBD embedded metadata capability enhances document provenance and asset authenticity and viable record keeping thus solidly future proofing the digital asset.

This Innovative leap for Content Management in the use of IFTBD metadata augmenting or replacing the established process of applying metadata within a digital asset management environment with entry and field constraints whatever the use or workflow is a new firm foundation in the science of metadata.

Share this Article:

Leave a Reply

Your email address will not be published. Required fields are marked *