Green, AI-based Video Compression

When:

Sunday, April 24 | 9:35 AM - 9:55 AM

Where:

W311-W313
Info Alert

Create or Log in to My Show Planner to see Videos and Resources.

Videos

Resources


{{video.title}}

Log in to your planner to join the zoom meeting!

Resources

Info Alert

This Session Has Not Started Yet

Be sure to come back after the session starts to have access to session resources.



Artificial intelligence (AI) is often seen as a way to introduce new system features and capabilities, at the expense of a significant increase in power consumption. Though, if the usage of heavy processing in an offline machine learning training phase leads to lower energy consumption during the deployment phase of a system, the carbon footprint can be improved at the end.

This presentation will present concrete AI-based techniques that can be applied to live video compression to reduce CPU consumption in the headend or to minimize the delivery and storage impact utilizing lower bitrates. It will also provide interoperability test results for each of the techniques discussed.


A Sub-Session Under:

Artificial Intelligence Applications for Broadcast

Speakers

Thierry Fautier
Vice President, Video Strategy
Harmonic

Session Type:
Paper Presentation