SciTransfer
EMERALD · Project

AI-Driven Automation to Reduce Energy and Costs in Media and Entertainment Production

digitalTestedTRL 5

Imagine if editing a movie or creating a virtual world took a fraction of the electricity and human effort it does now. This project builds smart tools that act like a high-speed assistant, automatically sorting through video footage and optimizing how data is processed. It's like upgrading from a gas-guzzling old truck to a streamlined electric vehicle for the digital content world.

By the numbers
30
project duration in months
7
consortium partners
57%
industry partner ratio
The business problem

What needed solving

The entertainment industry faces an unsustainable surge in video and XR content, leading to excessive energy consumption and a shortage of skilled human resources to process the data.

The solution

What was built

An open-source toolchain and API for measuring energy in media pipelines, and AI algorithms for automated player classification and action spotting in video.

Audience

Who needs this

VFX HousesStreaming Service ProvidersGame Development StudiosBroadcast EngineersMetaverse Content Creators
Business applications

Who can put this to work

Film and Television
mid-size
Target: Post-production house

If you are a post-production house dealing with massive amounts of raw footage and high electricity bills — this project developed AI-supported tools for automated video analysis and modification that increase production efficiency and use less energy.

Gaming and Metaverse
SME
Target: XR content creator

If you are an XR content creator dealing with unsustainable demands for data processing and skilled labor — this project developed more efficient data use for AI/ML and smart streaming to make media consumption more sustainable.

Broadcasting
enterprise
Target: Streaming platform

If you are a streaming platform dealing with high power demands for large-scale data processing — this project developed methods to exploit AV1 compression on general-purpose hardware to reduce energy consumption.

Frequently asked

Quick answers

What is the cost or pricing for these tools?

Based on available project data, specific pricing is not mentioned, but the project includes the creation of an open-source toolchain and API.

Can this be deployed at an industrial scale?

Yes, the project specifically targets large-scale media data processing and includes 4 industrial partners from the movie, broadcast, and streaming sectors.

What is the IP and licensing situation?

The project mentions the development of an open-source toolchain, API, and framework for quantifying energy consumption.

How does this integrate with existing software?

The tools are designed to measure energy consumption for commonly used processes, specifically mentioning integration with Foundry’s Nuke.

What is the implementation timeline?

The project is a 30-month initiative running from 2023-10-01 to 2026-03-31.

Consortium

Who built it

The consortium is heavily weighted toward commercial application, with 4 industry partners (57% of the group) and 1 SME, balanced by 2 universities. This structure, spanning 5 countries, suggests a strong focus on market-ready tools rather than pure academic research, specifically targeting the movie, broadcast, and streaming sectors.

How to reach the team

Contact UNIVERSIDAD POMPEU FABRA regarding the open-source toolchain and energy measurement API.

Next steps

Talk to the team behind this work.

Contact us to connect with the EMERALD consortium for early access to sustainable media processing tools.