SciTransfer
FITOPTIVIS · Project

Smart Video Processing That Runs Faster on Less Power — From Cloud to Edge

digitalTestedTRL 5
fitoptivis.eu

Imagine you have security cameras, medical scanners, and factory inspection systems all generating huge amounts of video — but the computers processing it are either too slow, too power-hungry, or too far away in the cloud. FitOptiVis built a way to split that video processing intelligently between local devices and remote servers, so you get fast results without burning through electricity. Think of it like a smart traffic controller for data — it figures out which parts to process locally and which to send to the cloud, optimizing for both speed and energy use. Philips Medical Systems led 30 partners across Europe to build and test this across real use cases.

By the numbers
30
consortium partners across Europe
5
countries involved (CZ, ES, FI, IT, NL)
16
industry partners in the consortium
11
SMEs participating
53%
industry ratio in consortium
16
total project deliverables
The business problem

What needed solving

Companies running video and image processing systems — from medical scanners to factory inspection lines to surveillance networks — face a constant trade-off: process locally for speed but burn energy, or send to the cloud for power but suffer latency. As camera resolution and sensor counts increase, this problem gets worse. There is no simple way to automatically optimize where and how video processing happens across a distributed system.

The solution

What was built

The project delivered a reference architecture for distributing image and video processing across edge devices and cloud infrastructure, along with optimization tools that balance performance against energy consumption at both design-time and run-time. Working demonstrator prototypes were built and tested against predefined requirements across multiple use cases, producing 16 deliverables in total.

Audience

Who needs this

Medical device manufacturers processing imaging data (CT, MRI, ultrasound)Factory operators using machine vision for quality control and defect detectionVideo surveillance companies managing distributed camera networksAutonomous vehicle developers needing low-latency image processingSmart city technology integrators handling real-time video analytics
Business applications

Who can put this to work

Medical Imaging & Healthcare Technology
enterprise
Target: Medical device manufacturers and hospital technology departments

If you are a medical imaging company dealing with slow image processing that delays diagnoses — this project developed an optimized pipeline that distributes processing between edge devices and cloud servers, cutting latency while keeping power consumption low. The system was coordinated by Philips Medical Systems with 30 partners testing across multiple use cases. With 11 SMEs in the consortium, the tools are designed for companies of varying sizes.

Industrial Quality Control & Manufacturing
mid-size
Target: Factory operators using machine vision for quality inspection

If you are a manufacturer relying on camera-based inspection systems that struggle with processing speed or energy costs — this project built multi-objective optimization tools that balance performance against power use in distributed vision systems. The consortium included 16 industry partners with real demonstrator prototypes tested against defined metrics. The reference architecture lets you plug in your existing cameras and processing hardware.

Smart Surveillance & Security
any
Target: Video surveillance integrators and smart city technology providers

If you are a security systems provider dealing with video feeds from dozens of distributed cameras that overwhelm your central processing — this project created a reference architecture for splitting video analysis between edge and cloud, optimizing for low latency. Tested across 5 countries with demonstrators validated against predefined requirements, the approach handles real-time processing at distributed sensor locations.

Frequently asked

Quick answers

What would it cost to implement this technology?

The project was a Research and Innovation Action, so the core tools and reference architecture were developed with public funding. Licensing terms would need to be negotiated with individual consortium members — Philips Medical Systems as coordinator or the 11 SME partners may offer commercial arrangements. Contact the consortium for pricing details.

Can this scale to industrial production environments?

The consortium included 16 industry partners and built demonstrator prototypes that were tested against predefined metrics and requirements. The multi-objective optimization approach was designed for real cyber-physical systems with distributed sensors and processing. Scaling to full production would depend on your specific hardware setup and processing volumes.

What is the IP situation — can I license this?

With 30 partners across 5 countries (CZ, ES, FI, IT, NL), IP is distributed across the consortium. Philips Medical Systems Nederland BV coordinated the project, and 16 industry partners hold various components. You would need to contact the relevant partner for licensing of specific tools or components.

How does this integrate with existing camera and processing systems?

FitOptiVis developed a reference architecture specifically designed for heterogeneous systems — meaning it works across different hardware and processing platforms. The approach covers the full pipeline from sensors at distributed locations through to cloud processing. Based on available project data, the architecture supports combined design-time and run-time optimization.

Is this ready for deployment or still at research stage?

The project produced demonstrator prototypes that were tested and compared against predefined requirements and metrics. With 16 industry partners including Philips Medical Systems leading, the technology has moved beyond lab research into validated prototypes. However, as a Research and Innovation Action that closed in 2021, additional engineering work may be needed for full commercial deployment.

What specific use cases were tested?

Based on available project data, the demonstrators covered multiple use cases involving image and video processing in cyber-physical systems. The deliverable descriptions confirm prototypes were built and results were assessed against requirements from WP1. Specific use case details would need to be obtained from the consortium or project website.

Consortium

Who built it

This is a heavyweight consortium with 30 partners, led by Philips Medical Systems — a global leader in medical imaging technology. The 53% industry ratio (16 out of 30 partners) signals strong commercial intent, not a purely academic exercise. With 11 SMEs alongside major industry players, the project blends scale with agility. The 12 universities provide deep research capability. The geographic spread across 5 countries (CZ, ES, FI, IT, NL) covers key European technology hubs. For a business looking to adopt this technology, the large and diverse consortium means multiple potential technology providers and integration partners to choose from.

How to reach the team

Philips Medical Systems Nederland BV (Netherlands) coordinated the project. Reach out via their corporate research partnerships division.

Order a report on this consortiumSee PHILIPS MEDICAL SYSTEMS NEDERLAND BV profile →
Next steps

Talk to the team behind this work.

Want to connect with the FitOptiVis team for licensing or integration? SciTransfer can arrange a direct introduction to the right partner in the consortium.

Order a report on this topicMore in Digital & ICT
More in Digital & ICT
See all Digital & ICT projects →
Affordable Satellite Communication Modules That Give SMEs Direct Access to Space Data
InnoSpaceComm
Open →
Home Dining Marketplace Connecting Travelers with Local Hosts Across Europe
ECINPERSON
Open →
AI-Powered 3D Art Creation That Cuts Production Time for Game and Film Studios
ArtomatixSuite
Open →