The LIST, a CEA TECH institute dedicated to digital technology, provides the industrial world with technological solutions enabling it to innovate in its production methods or to provide differentiation in its products. We cover the space between levels 2 and 7 of the TRL (Technology Readiness Level) scale, thus acting as a catalyst and accelerator of innovation for the company.

AI technologies for language and vision are of universal interest which leads us to intervene in most industrial fields and with the partnership mode suitable for each company, large group, ETI, SME and start-up.

We capitalize on these technologies, in the form of know-how and software, protected by numerous patents, property of the CEA and are implemented on our technological platforms, open to our partners.

Over the past 20 years, our commitment to these transfer processes has enabled us to develop a strong culture of research/industry collaboration, to validate our methodologies and to offer support adapted to each of our partners, extending when necessary after our technologies are put into production.

Semantic Analysis of Text and Images

We conduct research to better understand multimedia (text, image, speech, videos) and multilingual content. We filter, structure and organize the information in order to provide an easier access to the user.
We develop efficient and robust algorithms to analyze, classify and retrieve multimedia content and extract relevant information.

Computer Vision and Machine Learning for scene understanding

Our research in artificial intelligence also focuses on the design of new perception models for scene understanding, including algorithms for visual recognition in images and video sequences, the analysis of individual, group and crowd behavior, the characterization and intelligent annotation of visual data.

Vision for Modeling and Localization

Our activities in computer vision cover areas of research and applications such as geolocalization, precise object localization for augmented reality, robotic picking, conformity control, 3D reconstruction, diminished reality, realistic rendering and image enhancement, thanks to deep learning technologies.