Diminished reality for landscape video sequences with homographies

  • Authors:
  • Kosuke Takeda;Ryuuki Sakamoto

  • Affiliations:
  • Faculty of Systems Engineering, Wakayama University, Wakayama, Japan;Faculty of Systems Engineering, Wakayama University, Wakayama, Japan

  • Venue:
  • KES'10 Proceedings of the 14th international conference on Knowledge-based and intelligent information and engineering systems: Part III
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper describes a Diminished Reality application to eliminate occluders on videos capturing distant landscapes. Although many tourist spots have lookouts to see the distant landscape, the video capturing from there usually includes obstructive occluders such as pillar and muntin. The application vanishes the occluders in the video in semi-interactively with homographies of distant landscape among frames. For this purpose, the regions of occluders in every frame are segmented by the GrabCut algorism, and then the each region is recovered by the appropriate texture from another frame which is not occluded in same region.