Internship: 3D Luggage Volume Estimation and Owner-Luggage Association

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.

Purpose

This role is responsible for designing, developing, and improving algorithms that efficiently meet business and customer needs.

Airports are increasingly automating their processes at various stages of the traveler experience. In boarding areas, it is now possible to track and analyze travelers’ luggage in real time. For example, knowing the luggage volume in advance for each flight allows airlines to have sufficient time to efficiently manage emergencies when they arise. Similar functions are also useful for train stations and managing abandoned luggage on trains.

Key Missions

To achieve these goals, two main challenges must be addressed:

• Accurately estimating the 3D volume of luggage.
• Associating each piece of luggage with its owner, who will be identified later.

During the internship, the student will work on these two issues:

1. 3D Volume Estimation

Depending on the setup (single-camera or multi-camera), several approaches can be explored:

• Stereo vision
• Depth maps (e.g., DepthAnything models)
• 3D reconstruction (e.g., MoGE, SAM-3D)

2. Owner-Luggage Association

Possible strategies include:

• Associating pedestrian and luggage tracks using rule-based methods (track coherence, distance, proximity/configuration of key points, etc.)
• Using AI models that estimate relationships between objects (e.g., VSTRD, Open-VidVRD, Spatio-Temporal GNNs)

The intern will be responsible for reviewing existing literature to deepen understanding of one or both challenges, identifying relevant solutions, then implementing, training, testing, and analyzing the proposed solutions.

Profile & Other Information

Education

• Engineering student in final year of school
• Or Master’s 2 student specialized in deep learning and computer vision

Technical skills

• Strong knowledge of image processing, computer vision, and deep learning (preferably with PyTorch)
• Solid foundation in mathematics and data analysis
• Proficient in Python (C/C++ is a plus)
• Familiarity with image processing libraries is a plus

Personal skills and languages

• Curious, proactive, and autonomous
• Results- and solution-oriented
• Clear and convincing communication
• Fluent in English (spoken and written)
• Desire to contribute to the full algorithm lifecycle: design, implementation, integration, testing, and optimization