Artificial Intelligence Applications for Media Production

A Novel White-Balance System for Broadcast Cameras Using Machine Learning - $15

Color constancy is the ability to perceive colors of objects, invariant to the color of the light source. Our eyes, the human visual system is capable of doing that. Color constancy algorithms first estimate the color of the illuminant light source and then correct the image so that the corrected image appears to be taken under a canonical light source. In this project, we use color constancy on a broadcast camera without processing the video but instead acting on the RGB gain control much akin to a camera operator. The reference color is provided by a neural network which is trained to infer the color of the illuminant just looking at a sample frame from the camera; from that reference, we can compute the direction of the error from the white illuminant in color space and then generate a new RGB gain triplet to the camera. With the new adjustment, the cycle continues and eventually, the camera will output an image that is corrected and the neural network will estimate the illuminant as white yielding a zero error and the whole system stabilizes.

Edmundo Hoyle | GRUPO GLOBO | Rio de Janeiro, Brazil
Alvaro Antelo | GRUPO GLOBO | Rio de Janeiro, Brazil
Leandro Pena | GRUPO GLOBO | Rio de Janeiro, Brazil

Exploring the Creative Frontiers of AI for M&E Production and Distribution - $15

In recent years, Artificial Intelligence (AI) tools are beginning to revolutionize Media and Entertainment production and distribution industry. As with other industries, such as mobile, business, transportation, gaming, robotics, security, education.?Much like non-intelligent automation, artificial intelligence applications can be trained to perform human-like tasks. Some of the common skills currently imitated by AI include visual perception, speech recognition, decision-making, and adaptability

AI tools can be applied to perform tasks that were impossible to accomplish.?AI performs tasks faster and with greater precision than their experience human counterparts.?AI applications are beginning to augment the creative decision making skills of the production and distribution team, enabling energy, talent and enthusiasm to be focused on what only humans can do — at least for now.

During the M&E creative production process images, sound, and metadata are collected and stored, analyzed, identified and selected.?Recovery of content from the media libraries and archives is critical to the creative process.?This includes production generated multiple media formats of video, audio, graphics, and other types of data including text, scripts, and playlists.

It takes multiple tools to organize, collate and curate media and data from different formats and types. Metadata creation technologies help create the metadata necessary for content search and retrieval.?While several logging and recognition technologies can generate some metadata automatically, the results are often incomplete inaccurate and far from comprehensive.?

AI can be applied to myriad applications enhancing, accelerating and increasing the capacity of the media supply chain.?AI services such as transcription, language and dialect translation,?face and object recognition, social media sentiment analysis,?geolocation and fingerprinting are just a few examples.

Newly developed artificial intelligence for media asset management fuses the ability to understand all media and data types and perform discovery, synthesize and curate. ?

Across the media supply chain employing automated technologies that can create the metadata, then index and catalog it during the creation or acquisition process would help the creative process downstream.?There is a still a requirement for the creative team to identify relationships and the preferred clip selection.?It needs the legal team to add permissions, control access, contractual distribution and expiration criteria.?In news production, using machine-learning?techniques with pattern matching, additional content can be identified.?

This session will be rebroadcast on the?BEIT Express?channel?on May 15, 2020?at 12 a.m.?and 8?a.m.?EDT (UTC -4).

AI Closed Captioning by CaptionHub?

Gary Olson | GT Digital Ltd. | New York, NY, USA

The Podcast Promise: Topic Extraction for Monetization and Brand Targeting - $15

Audio content is having a renaissance. Of course, commercial broadcast, internet, and satellite radio are still important media, but podcasts ? those episodic bites of content that cover every topic under the sun ? are exploding. By some estimates, at least a million podcasts are being produced every year, and they?re reaching eager listeners on every type of smart phone, tablet, and computer as well as smart cars and ?connected home? platforms. But it?s not only the volume of podcasts that?s exploding, it?s also the variety of the content.

With so much content geared to many varieties of audiences, podcast producers are starting to realize the staggering potential for monetization. And likewise, direct-to-consumer advertisers are seeing dollar signs through a nearly untapped medium that is reaching virtually every demographic. The problem with a typical podcast, however, is that it?s a (sometimes hours-long) block of unstructured information. Until now, extracting content from a podcast that can be exploited with targeted advertising has been difficult if not impossible, since it has required human listeners to go through the program and tag specific topics. Also, by their nature, news, sports, and general interest podcasts are not ?long-tail? content and typically have a short shelf life, which has made topic extraction even less worth the effort. But new AI technologies are changing that equation.? ? Linear radio programmers can also use this technology to find interesting topics to curate and repurpose into new podcasts that will create even more inventory to cater to their audiences.

In this presentation, we?ll describe how new AI technologies are smoothing the way for brands to connect to their core audiences and immediately experience an uplift in engagement and revenues. Likewise, AI engines are? enabling content owners to monetize valuable content, even programs that appeal to super-niche audiences. By creating a powerful, multivariate search index to specific topics mentioned in the podcast, the AI tools enable brands to zero in on exact points in the program in which their products are likely to get the biggest reception. Take for instance, a podcast on homeopathic medicine and a brand that offers a homeopathic remedy for a specific condition. When the podcast reaches a specific point where that condition is mentioned, the AI platform tags that point for the ad buy. In other words, AI can connect the brand with consumers in exactly the right time and place in the podcast.

This session will be rebroadcast on the?BEIT Express?channel?on May 15, 2020?at 12:30 a.m.?and 8:30 a.m.?EDT (UTC -4).

Ryan Steelberg | Veritone, Inc. | Costa Mesa, California, United States