Studios Experiment With Metadata Capture During — Not After — Production

Image courtesy of the Tribeca Film Festival

The entertainment industry showcases some of its best work at the Tribeca Film Festival for the next two weeks, including the “The Suitcase,” a film by Abi Corbin about how the ordinary life of a Boston-bred baggage handler is turned upside down when he steals a suitcase that contains terrorist plans on 9/11.

It’s apt that “The Suitcase’s” premier coincides with the NAB SHOW, where the entertainment industry gathers to discuss how technology is changing how we consume media, because the making of “The Suitcase” explored this very issue.

While making “The Suitcase,” the Entertainment Technology Center at the University of Southern California (ETC) along with Disney|ABC Television Group and MarkLogic tested how creating metadata during movie production – instead of post-production – increased chances to serve movie lovers through technology while reducing costs for producers.

For producers, metadata can be as simple as noting locations. For example, when a scene is shot in Brooklyn, they would add a tag noting not only that the scene is in Brooklyn, but that Brooklyn is in New York City, and that the scene is about people standing on street corners during the heat of the summer. Fans or researchers do a simple search in the movie to find hang out places in New York. This ability gets more interesting when we think about finding funniest one liners, ultimate car chase scenes, or scenes with particular actors in them.

But there is potential to serve fans with even more detail. Imagine pausing a movie at a scene to find the backstory to it – why the director shot it that way, for example, or details on the scene, actors, and script.

“The Suitcase” project gave producers the chance to see the potential to do so by allowing them to capture descriptive metadata – information about what is happening on screen, including characters, locations, on-screen actions, story lines and scene breaks – when it is richest. Unfortunately metadata has been traditionally captured at the end – when rich context is forgotten.

A Better Fan Experience = Increased Revenue

The key was capturing everything about the movie during production so that producers thoroughly understand understood the content and how it’s it was produced. This in-depth, computer-stored knowledge allows them then to match content (snippets) to consumer profiles and behavior for massive customization, precise targeting and consumer engagement including:

• Creating different types of audience interactions
• Altering how viewers experience and interact with the content
• Facilitating dynamic advertising and recommendation insertions, based on user profiles and preferences
• On-target responses to key word searches
• And even creating versions of the original production, such as alternate endings

In short, preserving and cataloguing detailed data increases sales opportunities for content providers, helps develop new products, and promotes customer loyalty and satisfaction.

Production Metadata Cuts Cost of Movie Making

Tracking metadata also increases efficiency in making movies. As metadata is created and then used, producers, executives and others can view custom dashboards to evaluate production progress, monitor budgets and facilitate communication with the crew.

The particular details of what is included in the final version of the film are usually collected after production and after the final film edit. Adding metadata during production saves producers time as well as bypassing the expense and effort to find information for residuals and production costs after the film is made.

MarkLogic’s work with “The Suitcase” showed that linking actual shots to the final film edit generated the metadata producers needed to track this information rather than having to reproduce it after filming ended. The exact use case they explored was around location information. For example, when a new project is kicked off, researchers often look for scenes already shot in a particular location. Tracking this simple – and frequent – information is much easier as the movie is being made than when backfilling in post-production.

Although not done with the Suitcase project, the same is true with product placement. Producers usually track product placement by watching the movie after the final film edit, so tracking the information in the dailies also saves costs and increases revenue by increasing accuracy.

Getting the Data Back Together Again

Typically, a collection of independent companies, financiers and other parties collaborate to create a particular film, which means that a connected, robust infrastructure – from data management to software applications – is unlikely to exist.

This makes it difficult to get a holistic picture of costs, residuals, and product placements, especially when there may be different versions of data spread across many different parts of the organizations involved.

Data generated from the multiple systems that are used to run everything on set, from the cameras to the dailies, as well as the scripts, production accounting, physical assets and legal/contracts systems likely is scattered across multiple data stores each with their own IDs, making interoperability complicated, at best.

The result is disparate, siloed data systems that cause data loss, security issues and the need to recreate descriptive metadata of the film or television content after the production finishes. Not only is this inefficient, it is also a financial and time loss to the company distributing the content.

“The Suitcase” project targeted extracting these types of metadata during the production process, so that it did not have to be recreated downstream and using that data as an authoritative source for tracking production activities. The project ably illustrated that creating metadata during production is more time efficient and results in better data.

This new metadata management production workflow is implemented through the use of C4 – the Cinema Content Creation Cloud framework, and through the use of the MarkLogic Enterprise NoSQL multi-model database platform.

The project also demonstrated the benefits of creating and managing descriptive metadata with semantics or linked data for rapid, agile software development. MarkLogic created a search application for “The Suitcase,” that uses the tagged motion picture data mentioned earlier that lets users view the movie and the corresponding descriptive metadata at the same time. It is this application that sets the stage for custom information delivery by matching descriptive metadata with user appetites for information.

Not Reproducing Data Has Big Benefits

With new tools such as these on set, producers can generate rich metadata during production. This source data holds the promise to even better data driven experiences, reducing manual data management and rekeying, and better management of the entire production process including financial reporting. As more product windows are becoming custom with global theater releases followed by streaming and online releases sometimes only weeks later, it also makes the most of the, perhaps, scarcest commodity in the entire supply chain – time.

For More Information:

“The Suitcase” project will be featured in the May Issue of the SMPTE’s “Motion Imaging Journal” and was the subject of a session on broadcast infrastructure at SMPTE’s 2016 Technical Conference.