증강현실에서 3D이미지 구현을 위한 스테레오 정합 연구

The Study of Stereo Matching for 3D Image Implementation in Augmented Reality

  • 이용환 (극동대학교 스마트모바일학과) ;
  • 김영섭 (단국대학교 전자전기공학과) ;
  • 박인호 (단국대학교 전자전기공학과)
  • Lee, Yonghwan (Department of Smart Mobile, Far East University) ;
  • Kim, Youngseop (Department of Electronic and Electronical Engineering, Dankook University) ;
  • Park, Inho (Department of Electronic and Electronical Engineering, Dankook University)
  • 투고 : 2016.12.16
  • 심사 : 2016.12.26
  • 발행 : 2016.12.31

초록

3D technology is main factor in Augmented Reality. Depth map is essential to make cubic effect using 2d image. There are a lot of ways to construct Depth map. Among them, stereo matching is mainly used. This paper presents how to generate depth map using stereo matching. For stereo matching, existing Dynamic programming method is used. To make accurate stereo matching, High-Boost Filter is applied to preprocessing method. As a result, when depth map is generated, accuracy based on Ground Truth soared.

키워드

참고문헌

  1. Y.S. Kim, E.Y. Song, "The Algorithm of Brightness Control Disparity Matching in Stereoscopic", Journal of the Semiconductor & Display Technology, Vol.8, No.4, December, 2009.
  2. D. Scharstein, R. Szeliski, "A taxonomy and valuation of dense two-frame stereo correspondence algorithm", International journal of computer vision, Vol.47, No.1, April, 2002.
  3. M. Gong, Y.H. Yang, "Fast Unambiguous Stereo Matching Using Reliability-Based Dynamic Programing", IEEE Transactions on Patten analysis and machine intelligence, Vol.27, No.6, pp.998-1003, June, 2005. https://doi.org/10.1109/TPAMI.2005.120
  4. Y.S. Kim, E.Y. Song, "A Study of the Use of Step by Preprocessing and Dynamic Programming for the Exact Depth Map", Journal of the Semiconductor & Display Technology, Vol.9, No.3, September, 2010.
  5. Gonzalez and Woods, "Digital Image Processing", USA: Pearson Education, 2001.