lMAR: highly parallel architecture for markerless augmented reality in aircraft maintenance

  • Authors:
  • Andrea Caponio;Mauricio Hincapié;Eduardo González Mendivil

  • Affiliations:
  • Instituto Tecnológico y de Estudios Superiores de Monterrey, Monterrey, Nuevo León, Mexico;Instituto Tecnológico y de Estudios Superiores de Monterrey, Monterrey, Nuevo León, Mexico;Instituto Tecnológico y de Estudios Superiores de Monterrey, Monterrey, Nuevo León, Mexico

  • Venue:
  • Proceedings of the 2011 international conference on Virtual and mixed reality: new trends - Volume Part I
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

A novel architecture for real time performance marker-less augmented reality is introduced. The proposed framework consists of several steps: at first the image taken from a video feed is analyzed and corner points are extracted, labeled, filtered and tracked along subsequent pictures. Then an object recognition algorithm is executed and objects in the scene are recognized. Eventually, position and pose of the objects are given. Processing steps only rely on state of the art image processing algorithms and on smart analysis of their output. To guarantee real time performances, use of modern highly parallel graphic processing unit is anticipated and the architecture is designed to exploit heavy parallelization.