Idiap involved in a new European FP7 Project called "VANAHEIM"

VANAHEIM: Video/Audio Networked Surveillance System Enhancement through Human


  • (Coordinator) Multitel asbl, Belgium
  • Gruppo Torinese Trasportati, Italy
  • Idiap, Switzerland
  • INRIA, France
  • Department of Anthroplogy (Human Behavior Research group), University  of Vienna, Austria
  • RATP, France
  • Thales Communications, France
  • Thales Italia, Italy



The aim of VANAHEIM is to study "innovative surveillance components for autonomous monitoring of complex CCTV infrastructure" such as the ones prevalent in shopping malls or underground stations. VANAHEIM investigate three main application-driven research areas:

  1. "Scene activity modelling algorithms for automatic sensor selection". In CCTV system everyday practice, it is quite usual to see CCTV videowall monitors showing empty scenes, while there are obviously many cameras looking at scenes in which something (even normal) is happening. The development of tools to adaptively and autonomously select and propose the streams to be displayed in the control room is thus necessary. While this scenario is quite trivial when dealing with empty vs occupied scenes, building models to characterise the streams content in terms of (ab)normality turns out to be inescapable when dealing with almost all occupied scenes. Furthermore, the need for such selection is even more explicit when dealing with audio streams, for which mosaicing of data is not possible due to the transparent nature of sound.
  2. "Investigation of behavioural cues for human-centred monitoring and reporting." VANAHEIM will investigate the use of more subtle human behaviour cues (head pose, body pose) and social models (e.g. about space occupancy) to perform the live detection of well-defined scenario of interest for individual, group of people and crowd activities monitoring. VANAHEIM also targets to develop a situational awareness reporting subsystem, which could revolutionize the everyday operation of CCTV system, by displaying real-time user-oriented environmental figures, like approximate location and density of people and their behaviour (e.g. people walking, queuing at vending machine, …).
  3. "Collective behaviour building and online learning" " from long-term analysis of passenger activities." By combining cognitive science and ethological analysis, VANAHEIM targets the design of models able to identify and characterize the different structures inherent in collective human behaviour. In other words, by continuously analyzing, learning and clustering information about users’ locations, routes, activities, interactions with others passengers and/or equipments, and contextual data (time of day, density of people...), VANAHEIM targets the development of a system able to estimate the long-term dynamics and trends of large-scale human behaviour for the discovery of collective comprehensive daily routines.


The systems developed within VANAHEIM will be deployed and evaluated in both the Torino and Paris metros in order to demonstrate their scalability and performance.