Korean Journal of Optics and Photonics (한국광학회지)

Optical Society of Korea (한국광학회)
권호 표지
DOI QR코드

DOI QR Code

Object-Based Integral Imaging Depth Extraction Using Segmentation

영상 분할을 이용한 객체 기반 집적영상 깊이 추출

  • Kang, Jin-Mo (School of Electrical Engineering, Seoul National University) ;
  • Jung, Jae-Hyun (School of Electrical Engineering, Seoul National University) ;
  • Lee, Byoung-Ho (School of Electrical Engineering, Seoul National University) ;
  • Park, Jae-Hyeung (School of Electrical & Computer Engineering, Chungbuk National University)
  • 강진모 (서울대학교 전기공학부) ;
  • 정재현 (서울대학교 전기공학부) ;
  • 이병호 (서울대학교 전기공학부) ;
  • 박재형 (충북대학교 전기전자컴퓨터공학부)
  • Published : 2009.04.25
Download PDF
⟨ Previous Next ⟩

Abstract

A novel method for the reconstruction of 3D shape and texture from elemental images has been proposed. Using this method, we can estimate a full 3D polygonal model of objects with seamless triangulation. But in the triangulation process, all the objects are stitched. This generates phantom surfaces that bridge depth discontinuities between different objects. To solve this problem we need to connect points only within a single object. We adopt a segmentation process to this end. The entire process of the proposed method is as follows. First, the central pixel of each elemental image is computed to extract spatial position of objects by correspondence analysis. Second, the object points of central pixels from neighboring elemental images are projected onto a specific elemental image. Then, the center sub-image is segmented and each object is labeled. We used the normalized cut algorithm for segmentation of the center sub-image. To enhance the speed of segmentation we applied the watershed algorithm before the normalized cut. Using the segmentation results, the subdivision process is applied to pixels only within the same objects. The refined grid is filtered with median and Gaussian filters to improve reconstruction quality. Finally, each vertex is connected and an object-based triangular mesh is formed. We conducted experiments using real objects and verified our proposed method.

본 논문에서는 집적영상에서 깊이 추출을 할 때 영상 분할 방법을 이용하여 각각의 물체에 대해 삼각형 메쉬(mesh) 모델을 구성하는 방법을 제안하였다. 집적영상에서 렌즈 어레이와 카메라를 이용하여 실제 물체를 픽업하면 요소영상(Elemental image) 집합을 얻을 수 있다. 요소영상 집합은 3차원 물체의 정보를 가지고 있으므로 대응점 분석을 통해 깊이 추출을 할 수 있다. 우선, 각 요소영상 중심점의 대응점 분석을 통해 시차를 구하고 이를 이용하여 깊이를 구한다. 요소영상의 중심점에 해당하는 물체의 X, Y 공간좌표는 각 점들이 사각형 격자 형태를 이룬다. 이 격자 형태의 점들 중에서 가까운 점 3개를 연결하여 삼각형 메쉬를 만들면 물체의 삼각형 메쉬 모델을 구할 수 있다. 이 때 각 물체에 대해 삼각형 메쉬 모델을 구하기 위해서 요소영상의 중심점들로 구성된 가운데 방향별 영상을 영상 분할하고 각각의 분할된 영역에 대해서만 삼각형 메쉬 모델을 구성하였다. 영상 분할 방법은 normalized cut 방법을 이용하였다. 제안된 방법의 검증을 위해 실제 물체를 픽업하고 각 물체의 삼각형 메쉬 모델을 구성하였다.

Keywords

References

  1. G. Lippmann, 'Epreuves reversibles. Photographies integrale,' Comptes-Rendus de l'Academie des Sciences, vol. 146, no. 9, pp. 446-451, 1908
  2. J.-H. Park, S.-W. Min, S. Jung, and B. Lee, 'A new stereovision scheme using a camera and a lens array,' Conference on Algorithms and Systems for Optical Information Processing V, SPIE Annual Meeting, San Diego, CA, USA, Proc. SPIE, vol. 4471, pp. 73-80, Jul.-Aug. 2001 https://doi.org/10.1117/12.449359
  3. H. Choi, J.-H. Park, J. Hong, and B. Lee, 'An improved stereovision scheme using one camera and a composite lens,' Stereoscopic Displays and Virtual Reality Systems XI, Electronic Imaging, San Jose, CA, USA, Proc. SPIE, vol. 5291, pp. 196-203, Jan. 2004 https://doi.org/10.1117/12.526313
  4. J.-H. Park, S. Jung, H. Choi, Y. Kim, and B. Lee, 'Depth extraction by use of a rectangular lens array and onedimensional elemental image modification,' Appl. Opt., vol. 43, no. 25, pp. 4882-4895, 2004 https://doi.org/10.1364/AO.43.004882
  5. C. Wu, A. aggoun, M. McCormick, and S. Y. Kung, 'Depth extraction from unidirectional image using a modified multibaseline technique,' Stereoscopic Displays and Virtual Reality Systems IX, Proc. SPIE, vol. 4660, pp. 135-145, 2002 https://doi.org/10.1117/12.468026
  6. C. Wu, A. Aggoun, M. McCormick, and S. Y. Kung, 'Depth measurement from integral images through viewpoint image extraction and a modified multibaseline disparity analysis algorithm,' Journal of Electronic Imaging, vol. 14, no. 2, pp. 023018-1-023018-9, 2005 https://doi.org/10.1117/1.1902998
  7. C. Wu, M. McCormick, A. Aggoun, and S. Y. Kung, 'Depth mapping of integral images through viewpoint image extraction with a hybrid disparity analysis algorithm,' Journal of Display Technology, vol. 4, no.1, pp. 101-108, 2008 https://doi.org/10.1109/JDT.2007.904360
  8. G. Passalis, N. Sgouros, S. Athineos, and T. Theoharis, 'Enhanced reconstruction of three-dimensional shape and texture from integral photography images,' Appl. Opt., vol. 46, no. 22, pp. 5311-5320, 2007 https://doi.org/10.1364/AO.46.005311
  9. L. G. Shapiro and G. C. Stockman, Computer Vision (Prentice-Hall, New Jersey, USA, 2001), pp. 279-325
  10. J. Shi and J. Malik, 'Normalized cuts and image segmentation,' IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 22, no. 8, pp. 888-905, 2000 https://doi.org/10.1109/34.868688
  11. L. Vincent and P. Soille, 'Watersheds in digital spaces: an efficient algorithm based on immersion simulations,' IEEE Transactions of Pattern Analysis and Machine Intelligence, vol. 13, no. 6, pp. 583-598, 1991 https://doi.org/10.1109/34.87344