Image Composition Based on Blur Estimation from Captured Image for Augmented Reality

  • Authors:
  • Bunyo OKUMURA;Masayuki KANBARA;Naokazu YOKOYA

  • Affiliations:
  • Nara Institute of Science and Technology, Japan;Nara Institute of Science and Technology, Japan;Nara Institute of Science and Technology, Japan

  • Venue:
  • VR '06 Proceedings of the IEEE conference on Virtual Reality
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

Photometric registration is as important as geometric registration to generate a seamless augmented reality scene. Especially the difference in image quality between a real image and virtual objects decreases the seamlessness for users because the real image is degraded by blur effects caused by an optical system of a camera. To avoid this problem in video see-through augmented reality, it is necessary to simulate the optical system of the camera when virtual objects are rendered. This paper proposes an image composition method for augmented reality, which is based on blur estimation from a real image and rendering of virtual objects with the blur effect. Moreover, geometric registration is improved by estimating the camera position and orientation in consideration of the blur effect. In experiments, the effectiveness of the proposed method is confirmed by comparing a real image with a virtual object generated by our method.