How people do things?

Automatic capturing, recognition, analysis and rendering of human sensory-motor activities are essential technologies in many diverse applications, ranging from 3D virtual manuals through to training simulators and novel computer games.

In COGNITO, we have developed new capabilities of human activity capture, learning, analysis and monitoring. Moreover, we have linked these cognitive capabilities to a user interface, which allows context-sensitive assistance in complex industrial manual tasks.

The following video gives an overview of the COGNITO approach and single system functionalities.

The project approach is defined as follows:

  • To develop an on-body sensor network of inertial and inertial/visual sensors.
  • To infer osteo-articular biomechanical models of the body.
  • To capture user activity based on on-body cameras and inertial sensors.
  • To model and monitor the task space based on video data from on-body cameras.
  • To analyse user activity with both global ergonomic and musculoskeletal assessment.
  • To learn, analyse and monitor workflows.
  • To assist workflow execution through a context-sensitive user interface.

THEME ICT-2009.2.1 Cognitive Systens and Robotics
ICT – Information and Communications Technologies
Project Number: ICT - 248290
Date: 1.1.2010 - 31.12.2012
Project lead: Prof. Dr. Didier Stricker (Augmented Vision Group of the DFKI)