Sites Inria

Version française

Perception, Cognition, Interaction


The WILD project

In 2008, the Île-de-France Region and the Digiteo foundation  launched a call for projects to promote new research axes or enrich those already existing in major areas of concern (DIM), such as that of "Complex software and systems". The In-Situ research teams (Inria/LRI), Aviz (Inria) and AMI (LIMSI/CNRS) came together to answer this call.

© INRIA-Université Paris-Sud 11-CNRS / Équipe-projet IN-SITU

Named WILD (Wall-Sized Interaction with Large Datasets), the project proposes to create an interactive, high-resolution viewing platform unparalleled anywhere else in the world. Other platforms presenting a large display capacity exist. But for the first time, a platform has brought together a high display resolution, interactivity and collaborative functions. Several users can thus work on the same problem simultaneously, display and handle heterogeneous data, study different viewing angles of the same 3D scene,etc.


© INRIA-Université Paris-Sud 11-CNRS / Équipe-projet IN-SITU

The WILD platform is mainly aimed at the world of research. Indeed, researchers from across all disciplines are increasingly faced with the processing of data volumes that are both highly complex, large in size and heterogeneous (algorithms, images, etc.). The platform is made available to partner laboratories in order to work with them on their viewing needs.
The Applied Mathematics and Systems laboratory (MAS) with the Digiplante team Inria/École Centrale Paris/CIRAD) plans, for example, to use the wall to view, on different scale levels, the simulations of the major vegetation ecosystems produced with its growth models.

© INRIA-Université Paris-Sud 11-CNRS / Équipe-projet IN-SITU

Since the start of the WILD project, the platform has led to the opening-up of new research avenues in HMI, completed by the questions raised during participative design sessions conducted with the partner laboratories. Three essential points stand out:

  • How can one browse within very large images, data or documents, moving fluidly within an image on different scales, for example. This problem is clearly apparent, for example, in the case of viewing images of the sky, which are often large in size.
  • How can a large number of data be juxtaposed in order to compare them and highlight the differences between the components. Researchers at the Computer-Aided Neuroimaging Laboratory (LNAO)  by Neurospin (CEA) wish, for example, to be able to view a large number of brains all at once, in order to be able to compare them and group them according to their specific features.
  • How can they be made to collaborate, ensuring the coexistence of heterogeneous data, or to interact with multiple sources, possibly even remotely? A group of researchers needs to be able to work together, and understand what is happening simply and quickly.

The WILD project is thus only just beginning, and holds the promise of progress to be made both for the partner laboratories and within the field of HMI. Once again, the very difficulty lies in making interactions that are complex to begin with straightforward and natural. Mouse and keyboard are no longer enough; the platform will be used with intuitive objects adapted to each request… which remain to be discovered for the time being.

Keywords: Saclay - Île-de-France In-situ Interactive visualization Wall of images