Post-Doctoral Research Visit F/M Crowd dynamics data acquisition and processing for large-scale dataset construction

Post-Doctoral Research Visit F/M Crowd dynamics data acquisition and processing for large-scale dataset construction

Fonction : Post-Doctorant

The Inria Centre at Rennes University is one of Inria's nine centres and has more than thirty research teams. The Inria Centre is a major and recognized player in the field of digital sciences. It is at the heart of a rich R&D and innovation ecosystem: highly innovative PMEs, large industrial groups, competitiveness clusters, research and higher education players, laboratories of excellence, technological research institute, etc.

The VirtUs team at the Inria Centre at the University of Rennes is internationally recognized for its work in crowd simulation and the study of collective human behaviour. This postdoctoral position is part of the FOUL-X project (Programme Inria Quadrant), which aims to develop a new generation of crowd simulators capable of capturing the specific dynamics of crowds in real-world public spaces.

A key challenge in crowd simulation is the lack of datasets documenting the variety of crowd dynamics observed in different environments. Existing datasets are sparse and rarely capture the diversity of behaviours that emerge from different populations, activities, and spatial configurations. FOUL-X addresses this gap by designing and conducting field acquisition campaigns across multiple sites in France, with the goal of building an open, large-scale dataset of crowd dynamics.

This postdoc focuses on the data acquisition pipeline: from video capture in the field to the extraction of individual trajectories, and the characterization of crowd dynamics through dedicated metrics. The work will contribute directly to making the FOUL-X dataset available to the broader scientific community.

Assignments: With the help of the VirtUs team and under the supervision of Julien Pettré, the recruited person will be tasked with building a unique, open dataset documenting the diversity of crowd dynamics observed in real-world public spaces. This dataset will constitute a landmark contribution to the field, providing the scientific community with data capturing crowd behaviours across a variety of sites, populations, and spatial configurations - something that does not currently exist at this scale and diversity.

For a better knowledge of the proposed research subject: A state of the art, bibliography and scientific references are available on the VirtUs team website: https://www.inria.fr/en/virtus

Collaboration: The recruited person will work in close connection with a PhD student of the VirtUs team, who develops the video-based pedestrian tracking pipeline used to extract individual trajectories from field recordings, and with the second postdoctoral researcher of the FOUL-X project, who is responsible for the data-driven modelling activities. This triangular collaboration ensures that the dataset is built in direct response to both the technical constraints of the tracking pipeline and the scientific requirements of the modelling work.

Responsibilities: The person recruited is responsible for the design and execution of field acquisition campaigns across multiple sites in France, the validation and structuring of the resulting trajectory dataset, and the development of metrics to characterise and compare the diversity of observed crowd dynamics. The recruited person will take initiatives to maximise the scientific value of the dataset and ensure its open dissemination to the community.

Steering/Management: The person recruited will be in charge of coordinating field missions - including logistical, technical, and ethical aspects of data capture - and will lead the effort to make the FOUL-X dataset publicly available in a reusable and well-documented form

Phase 1 - Pipeline setup and campaign preparation (months 1-6)

• Evaluate and validate the video-based trajectory extraction pipeline developed by the PhD student of the team, with respect to crowd density, resolution constraints, and GDPR compliance requirements
• Define the minimal data resolution required to extract complete and accurate individual trajectories while ensuring data anonymisation
• Contribute to the identification and selection of acquisition sites, targeting a diversity of crowd dynamics (populations, spatial configurations, activities)
• Participate in the preparation of the ethical framework and site agreements for field data collection

Phase 2 - Field acquisition