Using the public internet for compressed, contribution-quality, real-time video transport has become commonplace for news, sports, and other real-time applications. Transport protocols are required to handle jitter and packet loss that can occur when using the internet. Some transport protocols are proprietary and closed, with no public information available about them. Other protocols have open specifications and offer open-source implementations. This paper presents a measured comparison between two open internet transport protocols: Reliable Internet Stream Transport (RIST) and Secure Reliable Transport (SRT). We used a network emulator to simulate impairments such as jitter, packet reordering, and loss, and compared the protocols using objective and subjective measures. As a reference, we also tested packet recovery using SMPTE ST 2022–1 Forward Error Correction (FEC), which was used for Internet contribution in prior decades before better protocols became available. The results show that both RIST and SRT perform well under low packet-loss scenarios, but RIST clearly outperforms SRT under challenging network conditions, such as those encountered with wireless IP links.
This February, I revised my travel schedule so that I would be in London for the fifth anniversary of the Young Innovators event, which I'm told was arguably the busiest and most energized (or, more appropriately, “energised”) yet! I had heard so much about this program over the years, and it sounded amazing. In fact, without ever having attended, our staff spoke with the Section about sharing the program with other Sections so that it could be reproduced in other parts of the world. And it did not disappoint.
The next generation of media engineers has been inspired by many great people and even more powerful events. The subject of this spotlight, Wren Wood, has been inspired by everyone from their father to the brave engineers working in active combat zones in Ukraine.
The European Broadcasting Union (EBU) Dynamic Media Facility (DMF) provides a layered model for software-defined production infrastructures running on commonly available IT infrastructure. The DMF seeks to enable customizable, dynamic workflows with cost-effective scalability and enhanced operational efficiency, security, and reliability to address the challenges of the evolving media business landscape. A key element of the DMF is the Media eXchange Layer (MXL), an open-source SDK for high-performance, live media interchange among containerized software media functions. MXL provides a vendor-neutral, interoperable platform with clear observability. MXL employs the lowest-latency asynchronous data transfer methods to minimize delays in the live signal chain. Shared memory is used for same-host functions, and remote direct memory access (RDMA) networking is used for cross-host communication. The Linux Foundation hosts the MXL Project in collaboration with the EBU and the North American Broadcasters Association (NABA). The MXL SDK code is available on a public GitHub repository.
Many years ago, I was watching the Winter Olympics on a CRT-based High-Definition TV in a development lab at work. ATSC was just becoming a standard, and I was astonished by the quality of the specular reflections in the ice during speed skating. I could not imagine what a 4K image would look like, let alone see it on a ${\$}$500 display. Nor could I imagine a day when millions of dollars' worth of live production gear would give way to some big, nebulous thing that everyone would call “the cloud.” Yet, here we are.
The Media & Entertainment industry is generating more data than ever before, yet much of it remains underutilized. Decades of centralized architectures, fragmented and siloed ownership, and tool-driven initiatives have produced data lakes and dashboards, but not the trusted, scalable insights the industry needs. At the same time, the rise of artificial intelligence has made the availability of high-quality, accessible data more critical than ever. This paper argues for a shift to a Data as a Product (DaaP) approach, supported by a Data Mesh architectural framework, as a path to unlocking value and securing the necessary trust. By treating data with the same discipline as software products, complete with ownership, lifecycle management, service-level agreements (SLAs), and feed-back loops, organizations can establish trust, reusability, and accountability. The paper outlines the three planes of a Data Mesh (infrastructure utility, mesh experience, and data product experience), defines what constitutes a true data product within this model, and highlights rights management as a representative use case where productization resolves longstanding pain points. Finally, the paper examines barriers to adoption and offers a road map to success. Benefits include faster insights, reduced costs, new monetization opportunities, AI readiness, and productivity gains. For Media & Entertainment companies facing rising complexity and competition, DaaP is no longer optional. It is a business imperative.
The April 2001Journal published in: “Balancing the Technologies in Digital Cinema Systems” by Steven. A. Morley, K. S. Thyagarajan, and Chris Irvine. “An efficient digital cinema system design balances the use of off-the-shelf technology with specialized hardware and software designed to meet the specific needs of this application. For instance, the use of standard computer, storage, networking, and communication equipment results in cost-effective design that benefits from other widespread uses of these technologies. However, certain functions required for a high-quality, reliable, secure digital cinema system must be custom developed specifically for this application. Electronic projection equipment, image compression systems, and security subsystems are three areas in which existing technology developed for other applications will not be adequate to meet the demands and current practices of cinema operations. This paper addresses the proper combination of off-the-shelf high-volume equipment and technology specifically developed for an acceptable digital cinema system solution.”
This paper proposes a distributed stream transformation architecture that integrates processing capabilities directly into IP-based media transport networks. As media distribution has transitioned from satellite to IP, solutions have been split between dedicated hardware and cloud platforms, each with scalability limitations. Our approach introduces a hardware-agnostic transformation framework supporting graphics processing unit (GPU), vision processing unit (VPU) (field-programmable gate array/application-specific integrated circuit (FPGA/ASIC)), and central processing unit (CPU) accelerators through unified Ku-bernetes orchestration, employing automatic repeat request (ARQ)-based protocols—Reliable Internet Stream Transport (RIST) and Secure Reliable Transport (SRT)—for reliable delivery over unmanaged networks. The architecture features a failure-tolerant control plane separated from the transport layer and a unified transformation engine running across high-performance core nodes and resource-constrained edge devices. Composable transformation pipelines eliminate duplicate processing by collocating multiple output requirements on optimal nodes. Cost analysis demonstrates that distributed commercial off-the-shelf (COTS) edge processing reduces per-stream costs from approximately ${\$}$281/month to ${\$}$97/month compared to managed cloud services over 36 months, supporting a hybrid deployment model.
Consumer mobile captures and professional movies are graded in different nominal ambient viewing environments, but many of them lack the corresponding Ambient Viewing Environment (AMVE) metadata. This causes problems in video rendering and degrades visual quality. This paper presents a recommended practice for handling AMVE metadata to benefit the industry and address practical issues. A theoretical model and a closed-form solution are provided, along with application examples and usage scenarios. We encourage the industry to explore this area further, align videos from different sources, and optimize the viewing experience.
The global news industry is undergoing a reshaping of its foundational processes as generative, multimodal, and agentic AI technologies permeate the entire content life-cycle—from acquisition and analysis to editorial development, regulatory compliance, production control, and multiplatform distribution. Previous technological transitions, such as the shift from film to tape, tape to digital, and later to cloud-driven IP workflows, primarily optimized discrete stages within the newsroom. The current Al-based transformation, by contrast, enables a fully integrated newsroom where all functional units are connected through a unified content backbone enriched with intelligent automation, continuous metadata generation, and real-time editorial support. This paper presents a consolidated framework detailing cautiously implemented approaches for specific functional units and theoretical/ futuristic implementations that synthesize earlier research on integrated newsrooms with the emerging paradigm of AI-assisted and agent-orchestrated editorial workflows. We describe how lightweight generative models, multimodal perception systems, and autonomous reasoning agents collaborate to accelerate ingest, shape story narratives, enhance fact-checking, perform frame-level compliance checks, automate virtual multicamera production, adapt content to multiple platforms, and generate localized headlines and graphics for diverse audiences. The framework further illustrates how AI-generated interactivity and personalization layers transform audience engagement, and how agentic AI can monitor and maintain operational stability across newsroom systems. The integrated model offers a scalable, future-ready approach that augments human editorial judgment, strengthens journalistic integrity, and increases overall operational efficiency.
The February 2001 Journal published in: “SMPTE Technology Committee on Digital Cinema-DC28: A Status Report” by R. M. Rast. “During the past year, a new SMPTE technology committee has been active, DC28, the Technology Committee on Digital Cinema. This report discusses its work and progress. - Most conclude digital cinema is the future of the consumer movie theater experience. Indeed, numerous press reports claim its imminence, and D-Cinema projection can be seen in several dozen theaters in the U.S. and abroad. However, what has been in the theaters has been in the nature of demonstrations of partial systems. DC28's activity applies to an overall end-to-end system for theatrical release of motion pictures, replacing the existing 35mm distribution model. The entire system must be defined, designed, proven in, and standardized. The work needed extends well beyond that which anyone organization, including SMPTE, can provide. However, SMPTE's standards work is essential, and it is now well underway.” For the full article, see: https://tinyurl.com/38km3ut6.
The broadcast industry's transition from Serial Digital Interface (SDI) to SMPTE ST 2110 IP infrastructure offers flexibility but introduces significant security threats. This paper addresses these challenges by proposing comprehensive best practice guidelines tailored to ST 2110 environments, leveraging the European Broadcasting Unio (EBU) Technology Pyramid for Media Nodes. We identify unique vulnerabilities, including the critical dependency on Precision Time Protocol (PTP) and the lack of native stream authentication. To mitigate these risks, the proposed framework integrates multilayered controls: robust network segmentation, secure Advanced Media Workflow Association (AMWA) Networked Media Open Specifications (NMOS) protocols, continuous monitoring, and device hardening. Furthermore, we explore the application of Zero Trust Architecture (ZTA) to reduce the attack surface in distributed workflows. This paper provides broadcast engineers with actionable technical guidance to evolve from traditional perimeter models to a modern Zero Trust approach, ensuring the benefits of IP media production are realized without compromising operational integrity.
As AI continues to dominate the conversation, the media tech industry is seeking knowledgeable leaders to implement what works, see through the hype, and teach professionals how to navigate these tremendous changes. Andy Beach is one such leader, navigating a ship through the seas of change and carrying the future of media technology with him.
Computer vision encompasses the analysis, processing, and interpretation of visual data. Tracking is a subset of this field, where systems recognize objects or salient features in a scene to determine their displacement across subsequent frames in a video stream. This facilitates automation, increases efficiency, and expands the functionality of these systems to applications in surveillance, medicine, and entertainment, among other fields. In recent years, Virtual Reality (VR) and Augmented Reality (AR) systems have gained popularity, prompting the development of camera tracking techniques. Camera tracking assesses the geometry and poses of a camera within a scene. Many tools are available to analyze and process camera tracking information, but most are proprietary, making information about them scarce; their availability to the general public also varies. To determine the democratization of the technology, three different tracking systems were compared. Two of these systems are standard tools used in the industry; the third system was a tracker built using OpenCV's open-source tools. A dataset of tracking values under different video parameters was gathered for all three trackers. By comparing and examining these results, it was determined that the tracking system was built using OpenCV and met industry standards. The impact of noise and lower resolution on the tracking system's performance was also assessed qualitatively by comparing tracking results in Unreal Engine. These results revealed that the democratization of tracking technology is limited by the equipment that the general public can access. This research aimed to understand better the workflow, optimization, and democratization of camera tracking systems.
The media and entertainment industry faces constant pressure to improve flexibility, increase efficiency, and “do more with less.” A Dynamic Media Facility (DMF)1 enables broadcasters and media companies to have precisely the technical resources they need-at the right time, in the right place, and with the correct configuration. In an ideal scenario, resource utilization rates approaching 100% can be achieved, a dramatic improvement compared to today's static, often underutilized broadcast infrastructure. To unlock the full potential of a DMF, every part of the operation must be digitized and orchestrated. Orchestration solutions are essential for managing the lifecycle of technical resources and integrating with broader service management and assurance workflows, including non-technical resources such as people or rooms, and business systems such as ticketing or billing. This paper explores the challenges of managing dynamic resources, examines dependencies such as compute, licensing, and network bandwidth, highlights what is already possible today, and outlines the remaining steps toward a fully dynamic, orchestrated media facility.
We present an automated shot-matching algorithm that incorporates perceptual effects and scene-specific context. Focusing on viewer-centric elements, such as attention to faces or objects and overall color mood, the system uses publicly available deep learning models to detect salient features. A committee-based approach matches these aspects independently, then combines them into a single RGB balance and black-level offset per shot. Although fully automated, the system supports manual adjustments to the strictness of face, object, mood, and temporal matching. This transparent, parameterized design avoids black-box behavior and provides intuitive, editable results for colorists. Visual examples demonstrate perceptual accuracy, and a study with five professional colorists evaluates match quality and potential time savings. Comparing manual grading of raw footage with algorithm-assisted pre-matched footage, results show a human-comparable match achieved 20% faster, with the algorithm reaching about 70% of colorist quality. These findings highlight the value of extended real-world evaluation.
his paper details CBC/Radio-Canada's ongoing journey into software-defined production workflows, a strategic initiative driven by our collaboration with the European Broadcasting Union (EBU) and other industry partners. By leveraging the Dynamic Media Facility (DMF) reference architecture, we aimed to tackle key industry challenges. We explore our approach to creating a universal user interface and offer an honest, real-world assessment of the latencies inherent in software-based systems. Our experience underscores the positive potential of the DMF model, highlighting our use of open-source initiatives and tools to close gaps.
2025 saw both continuing progress in media preservation and continuing challenges. Media preservation currently consists of two key, substantively different areas. The first step is to adapt current production tools and practices to facilitate the long-term preservation of new materials once the final production is completed. The second involves efforts to identify existing content libraries to create the best conditions for their survival and reuse. This includes planning for non-digital content masters (especially existing physical media) to be converted to digital assets, allowing for future reuse and remonetization.
The 2025 Media Technology Summit was held in a new location, featured new experiences, and attracted a larger audience amid global economic uncertainty. It was an unmitigated success that brought together the diverse membership of SMPTE and the professionals, students, and leaders who had yet to be introduced to the Society.
As this edition of the Motion Imaging Journal reflects on the progress our industry has made—and the developments yet to come—I find myself reflecting on the past year of my journey in this new (not so new anymore) role as Executive Director of SMPTE. It's hard to believe how quickly the time has passed. Most days, I catch myself thinking of something that happened six months ago as if it were just “last week.”
