DOI QR코드

DOI QR Code

Depth Map Generation Using Infocused and Defocused Images

초점 영상 및 비초점 영상으로부터 깊이맵을 생성하는 방법

  • Mahmoudpour, Saeed (Dept. of Computer and Communications Engineering, Kangwon National University) ;
  • Kim, Manbae (Dept. of Computer and Communications Engineering, Kangwon National University)
  • Received : 2014.01.21
  • Accepted : 2014.04.23
  • Published : 2014.05.30

Abstract

Blur variation caused by camera de-focusing provides a proper cue for depth estimation. Depth from Defocus (DFD) technique calculates the blur amount present in an image considering that blur amount is directly related to scene depth. Conventional DFD methods use two defocused images that might yield the low quality of an estimated depth map as well as a reconstructed infocused image. To solve this, a new DFD methodology based on infocused and defocused images is proposed in this paper. In the proposed method, the outcome of Subbaro's DFD is combined with a novel edge blur estimation method so that improved blur estimation can be achieved. In addition, a saliency map mitigates the ill-posed problem of blur estimation in the region with low intensity variation. For validating the feasibility of the proposed method, twenty image sets of infocused and defocused images with 2K FHD resolution were acquired from a camera with a focus control in the experiments. 3D stereoscopic image generated by an estimated depth map and an input infocused image could deliver the satisfactory 3D perception in terms of spatial depth perception of scene objects.

카메라 초점에 의해 발생하는 흐림(blur)의 변화는 깊이값을 측정하는데 사용한다. DFD(Depth from Defocus)는 깊이값과 흐림의 비례 관계를 이용하여 흐림의 양을 측정하는 기술이다. 기존 DFD 방법은 입력으로 두 장의 비초점 영상(defocused image)을 사용하는데, 기술적인 문제로 낮은 품질의 복원된 초점 영상(infocused image)과 깊이맵을 얻고 있다. 상기 문제점을 해결하는 방법으로 초점영상과 비초점 영상을 이용함으로써 복원된 초점 영상의 품질 저하를 해결한다. 제안 방법에서는 Subbaro가 제안한 DFD 방법에 새로운 에지 흐림 측정 방법을 결합하여 보다 정확한 흐림 값을 구한다. 또한 명암의 변화가 적은 영역에서는 흐림의 양을 측정하기가 어렵기 때문에, 관심맵(saliency)을 이용하여 비에지 영역을 채울 수 있도록 하였다. 실험에서는 초점 조절 기능이 있는 카메라로부터 20장의 2K FHD 해상도의 초점 및 비초점 영상을 생성한 후에 제안 방법을 이용하여 깊이맵을 생성하고, 마지막으로 입력 초점 영상과 깊이맵으로부터 3D 입체영상을 제작하였다. 3D 모니터로 시청한 결과 안정된 3D 공간감과 입체감을 얻을 수 있었다.

Keywords

References

  1. S. B. Gokturk, H. Yalcin, and C. Bamji, "A time-of-flight depth sensor," system description, issues and solutions," IEEE Workshop on Real-Time 3D Sensors and Their Use, 2004.
  2. P. Pentland, "A new sense for depth of field", IEEE Trans. Pattern Anal. Mach. Intel. pp 523-531, 1987.
  3. K. R. Ranipa and M. V. Joshi, "A Practical Approach for Depth Estimation and Image Restoration Using Defocus Cues", Workshop on Machine Learning for Signal Processing, 2011.
  4. Y. Wei, Z. Dong and C. Wu, "Global depth from defocus with fixed camera parameters," Proc. of IEEE ICMA, pp. 1887-1892, 2009.
  5. V. P. Namboodiri, S. Chaudhuri, and S. Hadop, "Regularized depth from defocus," Proc. of IEEE ICIP, pp. 1520-1523, Oct. 2008.
  6. H. Liu, Y, Jia, H. Cheng, and S. Wei, "Depth recovery from defocus images using total variation," Proc. of IEEE ICCMS, pp. 146-150, 2010.
  7. M. Subbaro, T. Wei and G. Surya, "Focused image recovery from two defocused images recorded with different camera settings," IEEE Trans. Image Processing, vol. 4, no.12, pp. 1613-1628, 1995. https://doi.org/10.1109/83.475512
  8. R. Achanta and S. Susstrunk, "Saliency detection for content-aware image resizing," Proc. IEEE ICIP, pp. 1005-1008, 2009.