Machine Learning Based UHD Up-conversion Using Generative Neural Networks - $15
Date: April 26, 2020Topics: 2020 BEITC Proceedings, Converting to Ultra HD Television BroadcastingUltra HD (UHD) is now part of mainstream TV production. Various consumer electronic companies now offer a wide range of high-quality UHD TV sets, which accounts for a significant proportion of all new TVs sold globally. The popularity of UHD has raised viewers’ expectations for a stunning quality experience. In today’s media landscape, online providers have moved quickly to satisfy this UHD demand via ABR delivery.
However, there remains a limited amount of content available that is UHD-native, and traditional content providers have a wealth of great content to offer the consumers even more value in UHD format. As consumer expectations grow, broadcasters will need to provide higher quality experiences, moving from a few UHD events to full-time or pop up channels, even if this is delivered only over ABR infrastructure. The question about where to get valuable UHD content remains open-ended, even though traditional up-conversion techniques can result in an end user experience that is more “HD-like”?than UHD.
This presentation explores a different approach to up-conversion, using Generative Adversarial Neural Networks to synthesize detail in the upconverted image, leading to an experience that is much closer to native UHD, leading to more compelling, higher quality experiences for consumers.
Alex Okell | MediaKind | Hedge End, Hampshire, United Kingdom
Tony Jones | MediaKind | Hedge End, Hampshire, United Kingdom
Machine Learning for Per-Title Encoding - $15
Date: April 26, 2020Topics: 2020 BEITC Proceedings, Improving OTT Video QualityVideo streaming content differs in terms of complexity and requires title-specific encoding settings to achieve a certain visual quality. Classic ?one-fits-all? encoding ladders ignore video-specific characteristics and apply the same encoding settings to all video files. In the worst-case scenario, this approach can lead to quality impairments, encoding artifacts or unnecessary large media files. A per-title encoding solution has the potential to significantly decrease the storage and delivery costs of video streams while improving the perceptual quality of the video. Traditional per-title encoding solutions typically require a large number of test encodes, resulting in high computation times and costs. In this paper, we illustrate a solution that implements the traditional per-title encoding approach and uses the resulting data for machine-learning-based improvements. By applying supervised, multivariate regression algorithms like Random Forest Regression, Multilayer Perceptron and Support Vector Regression we are able to predict mandatory video quality metrics (VMAF). That way, the test encodes are eliminated while preserving the benefits of per-title encoding.
Daniel Silhavy | Fraunhofer FOKUS | Berlin, Germany
Christopher Krauss | Fraunhofer FOKUS | Berlin, Germany
Anita Chen | Fraunhofer FOKUS | Berlin, Germany
Anh-Tu Nguyen | Fraunhofer FOKUS | Berlin, Germany
Stefan Arbanowski | Fraunhofer FOKUS | Berlin, Germany
Stephan Steglich | Fraunhofer FOKUS | Berlin, Germany
Louay Bassbouss | Fraunhofer FOKUS | Berlin, Germany
Making Your Assets Mean Something: Evolving Asset Management Systems Using Semantic Technologies - $15
Date: April 3, 2024Topics: 2024 BEITC Proceedings, Application of Large Language Model (LLM) in MediaProduce more content! Faster! Cheaper! Make it available in different places! And never throw anything away! It is seemingly impossible for media enterprises to keep up with these escalating and often conflicting business pressures. How can media asset management systems evolve to provide the right content based on the intent of the user? Can a creator produce content faster using knowledge hidden within the content? How can assets be easily accessed from distributed locations to enrich the experience? And does content growth exist without exploding costs? The answer to all these questions is yes. It is derived by leveraging tools that provide contextual meaning to the content, so that more effective utilization of the content can be performed. It goes beyond providing rich searching mechanisms to now providing the ability to get insights into the content. It relies on tapping into richer information sources that go beyond logged metadata for a piece of content, to now harvest knowledge from scripts, stories, and dialogue. The answer lies in creating a connected environment between data silos where a central brain has knowledge not just where content is located, but of what is in the content. The answer is in the ability for a Creator to use their native language and express in natural conversational form what they want to achieve when searching for, analyzing, and manipulating the content. The answer lies in tools understanding a user’s creative intent to perform appropriate actions, versus the creative having to learn and manipulate complex user interfaces to achieve the same intent. This paper discusses how certain artificial intelligence (AI) technologies such as Semantic Embeddings, Knowledge Graphs, and Multimodal Large Language Models are coming together under the umbrella of Knowledge Management to solve this problem of meaningfully extracting and using information from content that takes us beyond what we can do today with Asset Management systems.
Shailendra Mathur | Avid Technology | Montreal, Quebec, Canada
Rob Gonsalves | Avid Technology | Burlington, Mass., United States
Roger Sacilotto | Avid Technology | Burlington, Mass., United States
MC-IF VVC technical guidelines - $15
Date: April 3, 2024Topics: 2024 BEITC Proceedings, Video Encoding and CodecsVersatile Video Coding (VVC or H.266), standardized by ISO/IEC MPEG and ITU-T VCEG in 2020, offers best-in-class compression performance and has been selected (or is currently being considered) for use in next-generation broadcast and streaming standards around the world. These standards typically define VVC-based profiles and corresponding receiver capabilities. How the service is realized and the impact of the codec’s operational parameters on delivered compression performance is not in scope. The Media Coding Industry Forum (MC-IF) has developed – with input from broadcasters, encoder vendors and others in the community – technical guidelines that serve as a reference for VVC configuration choices to address operational, interoperability, and regulatory needs while achieving optimal compression performance. A community review phase concluded in late 2023, and the guidelines were made publicly and freely available at the start of 2024. This paper provides an overview of MC-IF VVC technical guidelines version 1.0. It gives an overview of the guidelines development process, scope, and contents. The paper provides a summary of VVC’s support in media transport and systems standards and follows with VVC adoption status in broadcast and streaming application specifications. A review of VVC-based profiles and receiver capabilities is presented. The paper concludes with an analysis on operational bitrates which can be expected with VVC for selected service deployment scenarios.
Lukasz Litwic | Ericsson | Gdańsk, Poland
Justin Ridge | Nokia Technologies | Dallas, Texas, United States
Alan Stein | InterDigital Communications, Inc. | Princeton, N.J., United States
Measurement of AM Band RF Noise Levels and Station Signal Attenuation - $15
Date: March 21, 2025Topics: 2025 BEITC Proceedings, AM Radio: Measurements and ModelingThis report covers measurements of RF noise levels on various roadway types from open interstate highways to city streets, to determine how the noise would affect AM broadcast reception. These environments reflect the current habits of AM radio listening, which is primarily in vehicles. In addition to RF noise level, RF signal levels were measured for three AM stations operating on frequencies in the lower, middle and upper portions of the AM broadcast band. These measurements provide a better understanding of how AM radio reception is affected by RF signal strength and noise in a range of roadway environments from rural to dense urban environments.
John Kean | Cavell Mertz & Associates | Alexandria, Va., United States
