NovelSense has won an EU funded project starting in February. In this project we are working on synthethic data for our traffic camera ABAKUS. The project named SYNTRA is funded within the EUHubs4Data, which is part of the Horizon2020 initiative to support innovative data-driven projects within the European Union.
Together with multiple partners, we are going to be working on the MVP for a SaaS platform which enables easy access to synthethic data for the training, testing and benchmarking of AI supported traffic cameras. This is relevant, as the acquisition of this data is currently difficult and costly. A reliable data source for GDPR compliant data sets could heavily boost the development and deployment of infrastructure systems using artificial intelligence. These can e.g. be used to generate reliable traffic data or support challenging maneuvers for automated and autonomous vehicles (e.g. at dangerous intersections).
Synthetic data sets can be a game changer for AI. The ability to create representative data sets which can include for example new types of vehicles (like E-Scooters or Cargo-Bikes) is very relevant. A key aspect of the project is the ability to validate and enrich the data sets by using our ABAKUS installations in the field. To learn more about the goals of project SYNTRA, visit the project site or contact me.
The project is done in cooperation with: Fraunhofer IOSB, Smart Data Innovation Lab (SDIL @KIT), Know Center (Austria), 4pdih (Slovenia).