/tng-media-pilot

5GTANGO Media pilot repository

Primary LanguagePython

Join the chat at https://gitter.im/sonata-nfv/Lobby

tng-media-pilot

5GTANGO Immersive Media pilot repository

The immersive media pilot aims to elaborate on the usability of 5GTANGO platform for media streaming applications for state-of-the-art technologies such as Virtual reality HMDs. Currently, the pilot focuses on development of components that use the 5GTANGO platform for streaming of single 360° camera footage with more secondary streams over the network with social media integration and automatic QoS detection to provide a seamless experience to the end user on a VR or non-VR device.

Folder structure

Folder Comment
emulators Source and Dockerfiles of the CPE/CCE
fsm FSM scripts
images images
k8s-deployments Deployment and service yaml for local service deployment
kubernetes Service descriptors for kubernetes deployment
mediapilot-service Service descriptors
video-test Service test descriptors
vnfs Source, Dockerfiles and other files of the VNFs

Use-cases description

Single location streaming

This scenario will work with the camera placed at the same location as the end user device. The camera can stream a video and the all the VNFs will be deployed at the same location.Depicts the VNFs that will be deployed for the NS to run.

Single Location Streaming

Multi location streaming

In this use-case, there would be multiple streams coming from the multiple locations to one or many users. Shows the deployment of VNFs at various locations for the NS. There would be single Content management VNF handling at least 20 users at once with multiple social media VNFs, Streaming servers deployed in the network service. The platform should be able to scale the VNfs based on the location requirements and the number of users.

Multi location streaming

Streaming with near real-time video analysis

In this use-case, there would be one main 360 stream which can have other supporting streams. The novelty of the use case would be the video analysis that will take place on the edge of the network in near real-time. The analysis will include game information, player statistics, highlighting of sports equipment such as balls in table tennis. Shows the usage of video analysis VNFs on top of UC1.

Streaming with Video Analysis

Network Service Components

  • Media-Content-Manager: The user (producers and clients) gateway to the service. Has an interface to register the other components of the service like cameras or streams.

  • Media-Aggregator: Receives streams from cameras and sends it to the streaming-engine or to another aggregator in a different node.

  • Media-Streaming-Engine: Implements the adaptive streaming functions. Transcodes the RTMP flow to different qualities (low, medium or source), finally serves the video to the clients.

  • Twitter-VNF:

  • YouTube-VNF:

Test components

  • Content-Producer-Emulator: Simulates the rol of cameras sending RTMP streams to the media-aggregator.

  • Content-Consumer-Emulator: Simulates the rol of the users playing the video from the streaming-engine.