Researchers at CEA-List worked with a team at CEA-IRFM to develop simulation and virtual reality technologies to evaluate part replacement scenarios.
Researchers at CEA-List brought their unique expertise to a new neural-network-based image analysis solution that successfully tracked soccer players in short video clips.
CEA-List provides strategic support to the Software Heritage Foundation, a multi-stakeholder nonprofit that is building a universal software archive.
CEA-List developed an AI-powered letter writing assistant and handed it off to France’s consumer protection agency.
CEA-List came up with the PREDIRE inspection system after being contacted by SARP for a solution in 2014. The smart wireless crawling drone is the ideal addition to the conventional—and more limited—video inspection techniques that have been used for a decade.
Neural network AIs are making inroads into autonomous vehicles, image and language processing, and a host of other use cases—some of which require a high degree of reliability. CEA-List developed a tool called PyRAT that ensures neural network AIs are robust enough to be implemented in critical systems.
CEA-List recently came up with a method for selecting the most suitable existing neural network for adaptation and reuse for new target applications. This advance makes it possible to train classification neural networks, or CNNs, without subject matter expertise or large datasets.
CEA-List has developed a new multimode method to improve the quality controls performed on the conductive inks used to make printed electronics. The goal is to help make sure the finished products operate as intended long after they are manufactured.