A Modular Vision Guided System for Tracking 3D Objects in Real-Time

  • Authors:
  • Markus Vincze;Minu Ayromlou;Wilfried Kubinger

  • Affiliations:
  • -;-;-

  • Venue:
  • CAIP '99 Proceedings of the 8th International Conference on Computer Analysis of Images and Patterns
  • Year:
  • 1999

Quantified Score

Hi-index 0.00

Visualization

Abstract

In industrial and commercial applications vision-based control of motion can only be feasible, if vision provides reliable control signals in real-time and when full system integration is achieved. A modular system architecture is built up around the geometrical features of the object. The initialisation is partly automated by using search functions to describe the task. The features found are then the basis for pose determination. Of particular need is a method of Edge-Projected Integration of Cues (EPIC) for robust feature tracking. A demonstration shows the robot following an object in 6 DOF with no special restrictions concerning room lighting, background are made.