About
The TEMPO project develops neuromorphic technologies & hardware, inspired by the human’s brain nerve network, to boost local data processing capability in edge AI devices. TEMPO broadens the applicability and strengthens the European ecosystem of integrated neuromorphic hardware.
This project accelerates the edge processing adoption in different application domains through a broad range of reference demonstrators.
These demonstrators validate the benefits and superior characteristics of the integrated neuromorphic technologies developed in the TEMPO project.
Concept and approach
The TEMPO project is based on collaborative research between world-leading research centres, large and small industrial enterprises, and universities.
As a strong team, the project activities focus on the following topics:
Emerging technologies
- Define emerging technology platforms
- Enable development through foundries
- Support development by systemized logistics
Technology architecture & design
- Design enablement
- Power-performance-area optimization
Algorithms and applications
- Identification of critical neuromorphic workload
- Application-driven DNN & SNN optimization
- Leverage technologies in application demonstrators
Facts & figures
2019May
Start
Application areas
Neuromorphic hardware technologies will be applicable in a broad range of use cases and applications. TEMPO proof-of-concept demonstrates their developed technologies over four application areas.
Health
Digital industry
Automotive
Digital life
TEMPO brings Europe in the pole position to address the neuromorphic market by strengthening the entire value chain in Europe.
Demonstrators
X-ray image denoising
https://bitcoin.org
Medical X-ray image denoising, increasing OR efficiency with improved imaging
Real-time nutrition assessment
Real-time nutrition assessment to obtain the caloric contents and nutritional values from food plate images
Human movement data analysis
Keyword spotting
Spotting keywords in speech, enabling next-gen voice-controlled virtual assistance
Vibration monitoring
Lidar-based object detection
Surround sensing for automotive, with enhanced energy efficiency, computing speed and smaller footprint
Video-based object detection
Object detection through versatile and flexible in-memory processing
Sound event monitoring
Sound localization through versatile and flexible in-memory processing
Radar-based object localization
Improved localization with radar to obtain enhanced angular positioning without increasing the the number of antennas
Gesture recognition
Gesture recognition at the edge, enabling intuitive & natural human-machine interaction
You can find the official descriptions and main outcomes of the TEMPO project on CORDIS, the project communication platform of the European Commission.
