Workshop on Visual and Inertial Sensing

Two-part special issue of
Journal of Robotic Systems

with InerVis 2003 papers

Workshop on

Integration of Vision and Inertial Sensors

29 June 2003

ICAR'03

ISR, Coimbra University, Portugal

Photo Album of INERVIS 2003

directory with hi_res photos (2048x1536)

ICAR 2003 Photo Album
directory with hi_res photos (2048x1536)

Introduction

Visual and inertial sensing are two sensory modalities that can be explored to give robust solutions on image segmentation and recovery of 3D structure from images, increasing the capabilities of robotic systems and enlarging the application potential of vision systems. Estimating the egomotion of an autonomous system is required in many important applications, e.g. navigation, 3D human-computer-interaction, and surveillance. Two sensing modes prove to be of particular value to achieve this task: visual and inertial sensing. The "beauty" of combining these two sensor modalities are the complementary characteristics of camera and inertial sensors. On one hand, the inertial sensors have large measurement uncertainty at slow motion and lower relative uncertainty at high velocities. Inertial sensors can measure very high velocities and accelerations. On the other hand, the cameras can track features very accurately at low velocities. With increasing velocity tracking is less accurate since the resolution must be reduced to obtain a larger tracking window with same pixel size and, hence, a higher tracking velocity.

In humans and in animals the vestibular system in the inner ear gives inertial information essential for navigation, orientation, body posture control and equilibrium. In humans this sensorial system is crucial for several visual tasks and head stabilisation. Neural interactions of human vision and vestibular system occur at a very early processing stage. The information provided by the vestibular system is used during the execution of visual movements such as gaze holding and tracking. Micromachining enabled the development of low-cost single chip inertial sensors. These can be easily incorporated alongside the camera imaging sensor, providing an artificial vestibular system. The noise level of these sensors is not suitable for inertial navigation systems, but their performance is similar to biological inertial sensors and can play a key role in artificial vision systems.

Scope

The goal of this workshop is to bring together researchers working on integrating these two and possible other sensors into one system. We encourage contributions including the following aspects:

Neurological studies of the human vision and inertial sensing,
Theoretical analysis of multi-rate signal integration,
Integration methods to fuse inertial, visual and other sensor modes,
Experimental results and evaluation of sensor integration techniques, and
Methods and metrics for performance evaluation.

Programme

Final Programme Leaflet

Attendance

Attendance Fees: Attendants are required to pay a small attendance fee (see Workshop/Tutorials Registration)
Travel Kit

Workshop site

The workshop will take place at the Institute of Systems and Robotics, at the Polo II University of Coimbra Campus.
Workshop BUS timetables and hotel map

Organizing Committee

Markus Vincze: ACIN, Vienna Univ. of Techn., Austria
Jorge Dias: ISR, University of Coimbra, Coimbra, Portugal
Peter Corke: CSIRO Manufacturing Science and Technology, Brisbane, Australia
Stefan Chroust: ACIN, Vienna Univ. of Techn., Austria
Jorge Lobo: ISR, University of Coimbra, Coimbra, Portugal

Program Committee

Ernst Dickmanns: Universität der Bundeswehr München, Germany
William Hoff:: The Colorado School of Mines, CO, USA
Axel Pinz: EMT, Graz University of Technology, Austria
Thierry Vieville: INRIA, Sophia Antipolis, France
François Berry: LASMEA, Université Blaise Pascal, France
Marnix Nuttin: Katholieke Universiteit Leuven, Belgium