Welcome

This is the home page of the Adaptive Video Delivery for Network Music Performance for Network Music Performance (AViD-NMP) project.

The AViD-NMP project is funded by the 6G-XR project.

The project is carried out by the Mobile Multimedia Laboratory (MMLab), one of the oldest research labs in Greece covering computing, communications and media. The MMLab is part of the Department of Informatics at the Athens University of Economics and Business (AUEB) in Greece. research projects are undertaken under the auspices of the AUEB Research Center (AUEB-RC).

About

The Adaptive Video Delivery for Network Music Performance for Network Music Performance (AViD-NMP) project will use the 6G-XR infrastructure to conduct experiments with Network Music Performance (NMP) applications over a 5G network, supporting immersive telepresence equipment and services such as holographic communications and avatar-based presence.
NMP can be a great showcase for both 5G and telepresence technologies.

Experiments

The following scenarios will be evaluated:

1. P2P with Native Rendering.
   • Sender encodes XR stream.
   • Stream is sent to each receiver independently.
   • Receiver(s) decode each stream locally.
   • Advantages:
     • Lowest possible latency
     • No need for MEC services
   • Disadvantages:
     • Sender needs to send multiple streams
     • Sender needs considerable processing power
2. Performance gains of edge-computing.
   • Sender encodes XR stream.
   • Stream is sent to SFU in MEC.
   • SFU relays the stream to each participant.
   • Receiver(s) decode each stream locally.
   • Advantages:
     • Sender only needs to send a single stream
   • Disadvantages:
     • Additional latency to go through MEC
     • Sender needs considerable processing power
3. SFU with Remote Rendering.
   • Sender does NOT encode the XR stream.
   • Stream is sent to SFU in MEC.
   • SFU encodes and relays the stream to each participant.
   • Receiver(s) decode each stream locally.
   • Advantages:
     • Sender only needs to send a single stream
     • Sender does not need processing power
     • SFU can adapt encoding depending on conditions
   • Disadvantages:
     • Additional latency to go through MEC

Objectives

• OBJ1: Compare the bandwidth, latency and processing requirements in the different experiment setups.
• OBJ2: Compare the QoE in the different experiment setups
• OBJ3: Assess the feasibility of XR-enabled remote music teaching
• OBJ4: Publish the results
• OBJ5: Release the developed code

News

• 23 April 2025 Project Started

Project Presentation