CrowdDNA is a collaborative project between 7 organizations from 4 European countries funded by the European Commission (EU H2020). It aims to enable a new generation of “crowd technologies”; a system that prevents deaths, minimizes discomfort and maximizes efficiency in the management of crowds.
This project has been selected in the FET Open call as a research and innovation program by the European Union Horizon 2020 (EU H2020) for the period of November 2020 to October 2024.

Technologies for computer-assisted crowd management
The CrowdDNA project aims to develop new technologies to help increase the level of comfort and safety of mass gatherings. The project’s vision for crowd technologies goes beyond all existing paradigms.
CrowdDNA relies on the 2 classical cornerstones of computer vision and simulation, but revolutionizes their combination to open a new era of efficient and robust crowd technologies that will rapidly become indispensable to equip any mass event.


An interdisciplinary approach
We bring together the academic expertise from cognitive sciences (ULM), biomechanics (Inria), physics (FZJ), computer science, including in particular Dynamic Modelling and Simulation (URJC), Signal Processing and Motion Analysis (UL), Crowd Modelling and Simulation (Inria).
The application aspects of the project are covered by two small ambitious companies: Onhys, a developer of crowd simulation software for the operational management of public places, and Crowd Dynamics International, a consulting firm for the organization of mass events and the design of infrastructures intended to welcome a large public.
Very promising results
Since the start of the project, a number of experiments have been carried out in small and large groups. A great deal of data was captured during these experiments. Some of this data is already available.
Several crowd observatories have also been installed at major events. The data collected during these events is currently being studied by the different partners.


Virtual all hands meeting and management board
CrowdDNA H2020 Project‘s partners had a productive week!

Safety science: Forward propagation of a push through a row of people
Sina Feldmann, Juliane Adrian

ICCV 2023: Hard No-Box Adversarial Attack on Skeleton-Based Human Action Recognition with Skeleton-Motion-Informed Gradient
Zhengzhi Lu, He Wang, Ziyi Chang, Guoan Yang, Hubert Shum

ArcTanGent Filming Summary and Experience
On the 17th of August CrowdDNA H2020 Project was running a Crowd Observatory at ArcTanGent festival!

Journal of Biomecanics: Stepping strategies of young adults undergoing sudden external perturbation from different directions
Thomas Chatagnon, Anne-Hélène Olivier, Ludovic Hoyet, Julien Pettré, Charles Pontonnier

Computer Graphics Forum: Model-based Crowd Behaviors in Human-solution Space
Wei Xiang, He Wang, Yuqing Zhang, Milo K. Yip and Xiaogang Jin