Pre-doctoral internship - Computer Vision - AI Engineering

Since our founding, IDEMIA has been on a mission to unlock the world and make it safer through our cutting-edge identity technologies. Our technology leadership makes us the partner of choice for hundreds of governments and thousands of enterprises in over 180 countries, including some of the biggest and most influential brands in the world. In applying our unique expertise in biometrics and cryptography, we enable our clients to unlock simpler and safer ways to pay, connect, access, identify, travel and protect public places – at scale and in total security. Our teams work from 5 continents and speak 100+ different languages. We strongly believe that our diversity is a key driver of innovation and performance.

The IDEMIA Research and Technology unit is responsible for developing new technologies to be included in all of IDEMIA's product lines, enabling augmented identity.

Composed of approximately 100 researchers and developers, the unit works on multiple application domains: facial recognition, fingerprint recognition, iris recognition, fraud detection, secure acquisition systems, video analysis, and document recognition and authentication.

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

This unit develops algorithms tailored to multiple use cases:

• Identity in civil applications (elections, biometric documents, etc.)
• Border control (fraud detection on documents or biometrics, facial recognition with comparison to the passport and reality, etc.)
• Police/Justice (criminal identification, object identification in videos, etc.)
• Access control (entry into a building or any secured area)
• Digitization of identity on smartphones (unlocking or payments using biometrics, opening a bank account remotely, etc.)
• Embedded terminals or smart cards (payment using biometric bank cards, etc.)

IDEMIA is recruiting excellent interns in Deep Learning and Computer Vision to enrich and strengthen its teams over a six-month period. The goal is to cover all aspects of scientific research, from state-of-the-art analysis, data collection and adaptation, modeling, and implementation, to the publication of articles or patents. The methods used include Bayesian methods, machine learning, deep learning (CNN, GAN, etc.), and all methods used in computer vision for still images and optimized automated video analysis.

This internship may lead to a PhD thesis.

Facial analysis and synthesis are both rapidly advancing fields. Among the areas of analysis, deep neural networks (trained through "deep learning") enable the quick estimation of semi-dense points of interest (or landmarks), typically numbering in the hundreds, on faces. These points provide an effective summary of facial geometry for numerous operations. The focus here is on improving the estimation of these points.

More specifically, during this internship, the goal is to implement 3D generative AI methods, particularly Neural Radiance Fields (NeRFs), to refine the estimation of landmarks. NeRFs construct coherent 3D representations in addition to producing high-quality 2D image synthesis, thus providing pseudo-annotations.

The intern will be tasked with conducting a literature review, implementing, testing, and improving the selected approaches. As part of this, they will:

• Design, implement, improve, and optimize statistically robust, efficient, secure, and scalable algorithms while respecting platform constraints.
• Adapt the algorithms to enhance the system and customer experience, create added value for clients, and succeed in external benchmarks.
• Develop test strategies and methodologies to evaluate new or competing algorithms.
• Present algorithms and results as part of internal community challenges.
• Stay up-to-date on the latest state-of-the-art research and algorithm development methodologies.
• Provide support for the design, implementation, customization, and operations of solutions.
• Close the loop with data/operations to ensure that the algorithms are fully effective in their operational use.

Engineering student from a top-tier institution in the final year of the engineering cycle (3rd year) or a Master’s student (M2), specializing in deep learning and computer vision.

Required Skills and Qualifications:

• Strong knowledge in image processing, computer vision, and deep learning (ideally with PyTorch).
• Solid background in mathematics and data analysis.
• Proficiency in Python programming.
• Experience with image processing libraries is a plus.
• Curious, proactive, and autonomous.
• Results- and solutions-oriented.
• Clear and persuasive communication skills.
• Strong proficiency in spoken and written English.
• Motivation to contribute to the full lifecycle of an algorithm: design, implementation, integration, testing, and optimization.