Reports

Tech Paper 2022: The paper presents an AI system for automated sports broadcast tagging, demonstrated with cricket.

Tech Paper 2022: This paper proposes an ultra-low bitrate video conferencing system using flexible avatar-based access patterns to save bandwidth, improve quality, and protect privacy.

Tech Paper 2022: This paper presents an Android TV-based sign language education app using AI and machine vision to help deaf children and their families communicate and learn together.

Tech Paper 2022: This paper explores encoding strategies for video game live streaming, focusing on game-specific content, compression optimizations, and real-time efficiency.

Tech Paper 2022: This paper presents a proactive smart home assistant designed to encourage sustainable behavior by raising energy awareness, suggesting actions, and adjusting appliances to support greener lifestyles.

Tech Papers 2021: NHK's "IoT-based Media Framework" aims to transition from traditional broadcasting to autonomous content presentation across various IoT devices. This framework utilizes semantic data about content, users, and their environment to create personalized, user-centric media experiences.

Tech Papers 2021: This paper outlines BBC R&D's Audio Orchestrator, an experimental tool that enables creators to rapidly produce immersive and interactive audio content by synchronizing multiple personal devices like phones, tablets, and smart speakers.

Tech Papers 2021: The paper proposes a framework for creating social controls in multi-user entertainment services, focusing on distributing authority and responsibility among individuals in close proximity.

Tech Papers 2021: The paper introduces Size-Aware Rate Adaptation (SARA) logic for the dash.js client, utilizing pre-encoded segment size information to enhance adaptive bitrate decisions. SARA aims to improve overall viewer experience by mitigating the impact of content-aware encoding's variable segment sizes.

Tech Papers 2021: This paper details a method using the Common Media Application Format (CMAF) with byte-range addressing in LL-HLS and LL-DASH to consolidate media objects for efficient cross-format, low-latency streaming, improving edge cache performance and reducing storage costs.