Utilizing Machine Learning for Video Processing

Liam Morrison

• The demand for media content in the form of audio, video, and images is growing at an unprecedented rate. To meet this demand, media content production is increasing rapidly. However, the process of producing, distributing, and monetizing content is often complex, expensive, and time consuming. Applying machine learning (ML) capabilities can allow for media companies to augment the work their teams are doing and provide capabilities that were previously prohibitively expensive or not even possible. ML capabilities are made available in one of three ways—built into software as a feature, consumed as an artificial intelligence (AI, of which ML is a subset) service [application programming interfaces (APIs) that can be consumed without ML expertise], and custom-trained models (which require data science expertise to train and deploy models)

Remote Innovations on the Cloud

Matt Herson 

• 2020 was a year of change in many ways. One of these aspects of change was the acceleration and adoption of remote workflows, leveraging collaboration tools for motion picture and television, born from the mandate to work from home virtually in every corner of the world. While COVID-19 had changed the industry in 2020, it accelerated moving these core production workloads to the cloud. Per Bob Chapek (CEO, Disney), “I would say COVID-19 accelerated the rate at which we made this transition, but this transition was going to happen anyway.” Cloud has helped customers navigate this new terrain, and we have witnessed technology providers and content creators collaborating like never before to push the art of the possible. People turned to entertainment to keep their spirits high, consuming content via streaming providers or broadcasters and the media and entertainment industry delivered. This is the story of a few companies that came together to contribute to that progress, adapting to the flexibility of modular design, while focusing on the end goal of a holistic post-production workflow anywhere in the world.

2021 High-Dynamic Range (HDR) Progress Report

Timo Kunkel, Patrick Griffis

• Since its introduction to the consumer market in 2015, high-dynamic range, better known by its acronym “HDR,” has established itself as a foundational component when looking at the aspects defining today’s image fidelity. This three-letter acronym “HDR” has become synonymous with brighter, more colorful pixels also known as (aka) “better pixels” enabled by a larger color palette that can better utilize the perceptual capability of human vision. SMPTE played a key role in enabling the HDR ecosystem with the standardization of an electro-optical transfer function (EOTF) based on the contrast sensitivity of the human visual system (HVS). This standard, ST 2084:2014, or more commonly “PQ” (for perceptual quantizer) establishes a practical maximum luminance range of 0–10,000 cd/m² for entertainment purposes much like 20 Hz to 20 kHz defines a practical frequency range for audio reproduction. Note that candelas/m² is a technical term for the perceived luminous intensity of light weighted by the HVS spectral response and is often colloquially referred to as “nits”—a term believed to come from the Latin word nitēre, “to shine.”

Standards

Bill Mandel and Bill Redmann

• This session explores what SMPTE has done in regards to standards for HDR in the past, how the Society influenced the industry and what is next. Industry experts will discuss HDR standards and technology and will answer questions.

American Society of Cinematographers Motion Imaging Technology Council Progress Report 2021 

Curtis Clark Chair, ASC Motion Imaging Technology Council (MITC)

Jesse Korosi, Greg Ciaccio, Jay Holben, Joachim Zell, Timothy Kang, David Morin and David Stump

• As we reflect on the disruptive impact that the pandemic has inflicted upon our personal and professional lives, I marvel at the resilience of our Motion Imaging Technology Council’s determination to continue addressing the rapidly evolving digital imaging technology challenges that are confronting filmmakers. Since its formation in 2002 as the ASC Technology Committee, our agenda over the years has consistently identified the critical technology developments driving the evolution of digital production workflows. The unique nature of our mission has always been the fusion of a solid scientific and engineering knowledge base with a laser focus on how these technologies can best benefit the cinematographer’s ability to utilize them to enhance their creative vision.