UPDATED 19:33 EDT / APRIL 10 2018

AI

Film and TV finally start to adapt to a new world where AI calls the shots

This month marks the 50-year anniversary of the world premiere of “2001: A Space Odyssey,” a film that captured a futuristic world where computers have the intelligence to control destiny. A half-century later, computers may not yet be in position to control our lives completely, but they are poised to make the very movie that predicted it.

In particular, artificial intelligence is working its way into the film and television industry tool box. What that means for the industry is the center of discussion this week at the National Association of Broadcasters Show in Las Vegas as a parade of technology companies, including IBM Corp., Amazon Web Services Inc. and Intel Corp., participated in the gathering to offer a glimpse of media’s future.

“We work with the storytellers and make things possible that years ago you could only read in a novel,” Lynn Comp, vice president of Intel’s Data Center Group and general manager of its Visual Cloud Division, said during a keynote session on Monday.

IBM Watson at the Masters

When it comes to artificial intelligence, the book’s first chapter is already written. IBM’s Watson Media, formed only nine months ago, has landed high-profile projects to provide AI support for big-tent events such as the U.S. Open tennis tournament, music’s Grammy Awards and the recently concluded Masters golf event.

At the Masters, IBM handled the livestream and used AI to select highlights of the action. The technology combines audio cues such as a sudden roar of the crowd with recognized visual images such as a fist pump to identify highlights from the day’s action.

“The impact of AI can be so great across all of the businesses here,” David Mowrey, head of product and development at IBM Watson Media, said in an exclusive interview with SiliconANGLE. “It’s really catching up fast.”

The advancing ability for AI to recognize visual images and act swiftly with accuracy goes beyond the world of sports and entertainment. Broadcasters and film producers are drowning in a sea of content, and the ability to search and discover the right clips rapidly in the proper context could become a major game changer.

To that point, Mowrey demonstrated IBM Watson’s technology applied to TED Talks, the technology forum begun in 1984 that has amassed a vast archive of recorded presentations by global leaders. A search under the word “happiness” instantly generated a ranked list of videos tagged at the precise point where the word is mentioned by a speaker.

“Being able to derive value from an archive is something we hear all of the time from our clients,” Mowrey said. “The more you know about a video, the better recommendation you can provide.”

Automating content production

Other companies are using AI to prepare content for airing. Amagi Media Labs has launched a machine learning powered content preparation suite called Tornado, which is designed to process ad break points, color correction, noise reduction and insert credits without any human involvement. Much of Amagi’s machine learning work is done on the AWS platform using the open-source deep-learning software framework TensorFlow.

“Machines are already helping the content production process,” Baskar Subramanian, co-founder of Amagi, said during a presentation on Monday at NAB. “It’s largely automating labor-intensive jobs.”

Behind the logistics of media production lies a tantalizing view of future use where AI and machine learning will be able not only to extract action from raw metadata, but to deduce style and genre as well. Cutting highlights and inserting ads are all fine, but the media industry is captivated by the potential to personalize the content in ways that all humans can appreciate.

“Can I buy an Akira Kurosawa machine model for filmmaking?” mused Sabramanian, in reference to one of the most influential filmmakers in the history of cinema. “It’s a rich field and a lot of possibilities are happening.”

Amazon targets media workflow

AWS is seeking to position itself at the intersection between machine learning and media production. The company set up a major pavilion (pictured) in the center of the NAB conference this week to demonstrate its AI and machine learning capabilities.

“AWS takes machine learning services and incorporates them into the day-to-day media workflow,” Usman Shakeel, worldwide technology leader, media and entertainment at AWS, said during an NAB presentation today. “All you have to do is bring your data.”

Shakeel briefly described media use cases where AWS machine learning tools are being applied for a variety of purposes. C-SPAN uploaded over 200,000 hours of video and used the AWS Rekognition cloud visual analysis service to create a searchable database. Conde Nast has applied machine learning to improve the user experience on its website, where the timing and placement of stories are controlled by AWS algorithms.

Despite advances and examples such as these, Shakeel cautioned that the application of machine learning in the media world is still a work in progress. “There is no Swiss Army knife when it comes to machine learning in content production,” Shakeel said. “Most of these platforms are still evolving.”

While AI will help the media industry fulfill its role as global storytellers, it’s still going to take a big, fast pipe to process and distribute all of that content. The coming of fifth-generation wireless systems or 5G is viewed by industry leaders as a key element in the implementation of new media production tools and delivery systems.

Speed and capacity, particularly as intelligence tools and other applications move to the edge, will be especially crucial. “With 5G you’ll have up to 10 times less latency, and you can support up to 100 times more users,” said Intel’s Comp. “That technology combined with moving to the edge is really going to help with storytelling.”

At one point in “2001: A Space Odyssey,” during a tense exchange between a space astronaut and the onboard computer Hal, the dialogue takes a suddenly stubborn tone. When Hal refuses to open a portal, it says, “I think you know what the problem is just as well as I do. This mission is too important for me to allow you to jeopardize it.” If today’s mission is to integrate AI, Hal’s film and TV colleagues apparently feel much the same way.

Photo: Mark Albertson/SiliconANGLE

A message from John Furrier, co-founder of SiliconANGLE:

Your vote of support is important to us and it helps us keep the content FREE.

One click below supports our mission to provide free, deep, and relevant content.  

Join our community on YouTube

Join the community that includes more than 15,000 #CubeAlumni experts, including Amazon.com CEO Andy Jassy, Dell Technologies founder and CEO Michael Dell, Intel CEO Pat Gelsinger, and many more luminaries and experts.

“TheCUBE is an important partner to the industry. You guys really are a part of our events and we really appreciate you coming and I know people appreciate the content you create as well” – Andy Jassy

THANK YOU