Hybrid Research/AI Engineering

Made up of around 100 researchers and developers, the unit works on multiple application areas: facial recognition, fingerprint recognition, iris recognition, fraud detection, secure acquisition systems, video analysis, document recognition and authentication.

All these technologies are developed to achieve the world's No. 1 rank in their category, guaranteeing a high level of security, while improving the comfort of the user experience for the end user.

This unit develops algorithms adapted to multiple use cases:

• Identity in the context of civil applications (elections, biometric documents, etc.)
• Border control (Fraud detection on documents or biometrics, facial recognition with comparison with passport and reality...)
• Police / Justice (identification of criminals, identification of objects in videos...)
• Access control (entry into a building or any secure area)
• Road safety (radar equipped with video technologies, etc.)

IDEMIA is recruiting excellent researchers to enrich and strengthen its teams. The aim is to cover all aspects of scientific research, from data collection and adaptation, modeling, implementation to the publication of articles or patents. The methods used are Bayesian methods, machine learning, deep learning (CNN, Transformer, GAN, Diffusion models...), and all methods used in computer vision for still images and optimized automated video analysis.

• The study and development of new algorithms at the cutting edge of technology
• Optimization of algorithms, depending on use cases
• Performance measurement and analysis, including fraud
• Technological monitoring and analysis of the state of the art of research
• The expertise to analyze the problems encountered in the field, and win benchmarks
• Support to operational teams to define the best personalized solutions for certain specific clients
• Integration into environments as diverse as professional servers in the cloud or on site, embedded terminals, or smartphones
• The areas of application will be the detection of faces in images and videos, video analysis in general as well as the development of freeflow algorithms (tracking and recognition of people in a video stream)

We welcome people with different degrees of experience and skills. Our team is constantly learning and experimenting, making it an ideal place for juniors eager to learn or seniors looking for technical challenges.

• Theses or engineering school
• First significant experience in image processing, computer vision, deep learning (ideally with pytorch)
• Solid training in mathematics, data analysis, C, Python, pytorch, panda, ...
• Ability to process a large amount of data and work in a research environment, i.e. in an unknown field
• Curious
• Proactive, autonomous
• Oriented towards results and solutions
• Clear and convincing communication
• Fluent spoken and written English and French
• Additional language would be appreciated