Internship: Tracking for Crowded Freeflow Environments

Purpose

Context:

Within IDEMIA, the Research & Technology unit plays a key role in developing the technologies of tomorrow, which are integrated across all our product lines.

Comprising around 100 passionate researchers and engineers, the team pushes the boundaries of innovation every day in high-impact domains such as facial recognition, fingerprint and iris recognition, fraud detection, secure acquisition, video analysis, and document recognition and authentication.

This internship focuses on multi-object tracking in freeflow environments, where objects move freely without physical constraints. Tracking in such crowded scenes is challenging due to frequent occlusions that often lead to identity switches or tracking loss. The objective is to develop a robust and reliable state-of-the-art tracking system capable of handling these complex freeflow scenarios. This research is of relevance to Idemia, as it directly supports applications in freeflow identification and intelligent surveillance.  

 The internship will last 6 months, starting in March 2026, and will be based in Courbevoie (La Défense). 

Key Missions

• Several research directions can be explored to address the challenge, including: 

- Leveraging Amodal Multi-Object Tracking and Segmentation, to reason beyond visible regions; 

- Exploiting Re-Identification (Re-ID) datasets with modern tracking architectures (e.g., MOTIP); 

- Incorporating 3D information into multi-camera setups to exploit spatial consistency and cross-view tracking. 

• Key responsibilities include: 

- Analyzing and understanding the current tracking model architecture and performance. 

- Conducting experiments to evaluate tracking performance in relevant benchmarks. 

- Collaborating with a team of researchers to contribute to the overall project goals. 

- Documenting findings and presenting results to stakeholders. 

Profile & Other Information

Education:

• Engineering student in the final year of the engineering cycle, or a Master's student (M2) specializing in computer vision, image processing, or Deep Learning. 

Required Technical Skills: 

• Strong knowledge of computer vision and Deep Learning. 
• Proficiency in Python, OpenCV, and PyTorch (or similar frameworks). 
• Solid training in data analysis and software development. 

Interpersonal Skills and Languages:

• Proficient in English, both spoken and written (e.g., reading scientific articles in English, presenting work). 
• Curious, proactive, and autonomous. 
• Results-oriented and solution-focused. 
• Clear and persuasive communication skills. 
• Eager to contribute across the algorithm lifecycle: design, implementation, integration, testing, and optimization.