Tech Papers 2023: This paper details a proof of concept that optimizes the Virtual Reality (VR) streaming experience by blending client-side Quality of Experience (QoE) data with real-time network analytics to accurately diagnose and resolve performance bottlenecks.
Tech Papers 2023: This paper explores how MetaHumans and digital doubles are changing sports broadcasts today, and the possibilities that they’ll bring to future match day coverage.
Tech Papers 2023: This paper introduces the MPEG-I Scene Description (MPEG-SD) framework (ISO/IEC 23090-14) as a standardized, dynamic, and interoperable solution for composing, delivering, and enabling rich interactivity within real-time immersive 3D media and hybrid environments
Tech Papers 2023: This paper details a subjective study that found current film grain synthesis methods in video codecs, such as AV1, are suboptimal and fail to consistently preserve the filmmaker's creative intent, highlighting the need for a better objective quality metric.
Tech Papers 2023: This paper provides a technical overview of state-of-the-art deep learning models for video super-resolution (SR) enhancement, such as PSNR-oriented methods, Generative Adversarial Networks (GANs), Transformer Models, Flow Models, and Diffusion Models, and compares their performance and computational complexity for use in cloud-based UHD broadcast up-conversion.
Tech Papers 2023: This paper demonstrates how the Versatile Video Coding (VVC) standard enables the use of more efficient open Group of Pictures (GOP) coding in adaptive streaming workflows, significantly improving compression efficiency and subjective video quality compared to the legacy closed GOP method.
Tech Papers 2023: This paper discusses VVC/H.266, the newest video coding standard designed to significantly improve compression efficiency over HEVC/H.265 and provide efficient coding for a broader range of video formats.
Tech Papers 2023: This paper explores how NHK Enterprises, Inc. has produced an avatar, KIKI, that looks like a real person.
Tech Papers 2023: This paper outlines a survey of minute-by-minute audience consumption of radio and television programmes for 3-4 days post-broadcast, spanning a period of twelve months
Tech Papers 2023: This paper explains a machine learning (ML) system that analyzes the visuals, audio, and transcript of a video segment to determine its dominant emotion and then select or generate the most aesthetically fitting background music to reduce production time and cost.
