Abdelmajid Moussaoui - ATEME
Kristen Grauman - University of Texas at Austin and Facebook AI Research
RTP Payload Format for JPEG XS
Speaker: Thomas Edwards
- JPEG XS is a low-latency, low-complexity wavelet codec that is promising for IP transport of professional media both on-premises and on the cloud. A combination of standards and specifications, including SMPTE ST 2110-22, ISO/IEC 21122-3, IETF RFC 9134, and VSF TR-08 define the transport of JPEG XS in RTP over IP.
Dynamic Seamless Resource Allocation For Live Video Compression On A Kubernetes Cluster
Speakers: Abdelmajid Moussa and Thomas Guionnet
- A solution is proposed on top of Kubernetes to dynamically allocate services resources without service interruption. It serves as the basis for optimizing a live video compression service. It is demonstrated that dynamic resource allocation can benefit to a video compression application, either by reducing the resource consumption, hence costs, or by enhancing delivered video quality. By combining the proposed solution with an elastic encoder and machine learning for content complexity estimation, a content and application aware dynamic resource orchestrator for real-time video compression is designed. Preliminary experimental results using ATEME Titan Live Micro-services encoders demonstrate up to 9.4% average bitrate reduction and up to 12.8% bitrate reduction on the most demanding channel.
Teaching AI to perceive the world through your eyes - Facebook
Speaker: Kristen Grauman
- AI that understands the world from a first-person point of view could unlock a new era of immersive experiences, as devices like augmented reality (AR) glasses and virtual reality (VR) headsets become as useful in everyday life as smartphones. Imagine your AR device displaying exactly how to hold the sticks during a drum lesson, guiding you through a recipe, helping you find your lost keys, or recalling memories as holograms that come to life in front of you.