효율적인 깊이 특징 추출을 이용한 스테레오 영상 기반의 3차원 모델링 기법

Stereo Image-based 3D Modelling Algorithm through Efficient Extraction of Depth Feature

  • 하용수 (금오공과대학교 컴퓨터공학과) ;
  • 이행석 (금오공과대학교 컴퓨터공학과) ;
  • 한규필 (금오공과대학교 컴퓨터공학과)
  • 발행 : 2005.10.01

초록

본 논문에서는 특징 기반 3D 모델링 알고리즘을 제안한다. 깊이 기반 기술을 다루는 전통적인 방법들은 영상 정합을 위한 깊이정보추출에 많은 시간을 필요로 한다. 특징 기반 알고리즘에서 삼각형 내부의 모든 픽셀들에 대한 모델링 오차 계산이 필요하다 할지라도 깊이 기반 보다는 특징기반 방법들이 보다 적은 계산 부담을 가지나 이는 또한 계산 시간을 증가 시킨다. 그러므로 제안된 알고리즘은 효율적인 3D 모델을 생성하기 위해 초기 3D 모델 생성, 모델 평가 및 모델 세분화의 3단계로 구성하였다. 초기 모델 생성을 위해 자기 변화와 델루니 삼각화가 사용되었고 이 단계에서 빠른 경계 추출과 점진적인 델루니 삼각화 및 삼각형 내부의 중심에 가까운 정점을 선택하거나 모든 픽셀에 대한 오차 계산을 위한 연산 시간을 줄이기 위해 형태학적 미분 연산자를 수정하여 이용하였다. 모델 생성 후 평가 단계에서 표면의 변이 변화와 근사 오차 및 표면의 크기를 평가하여 드물게 정합을 수행 하였고, 그 후 큰 오차를 갖는 표면들을 선택하여 작은 표면이 되게 세밀화 작업을 했다. 실험 결과 제안된 알고리즘이 평탄영역 및 급격한 영역에서 보다 적은 모델링 오류로 적응적인 모델을 획득할 수 있었고 모델 획득시간을 현저하게 줄일 수 있었다.

A feature-based 3D modeling algorithm is presented in this paper. Since conventional methods use depth-based techniques, they need much time for the image matching to extract depth information. Even feature-based methods have less computation load than that of depth-based ones, the calculation of modeling error about whole pixels within a triangle is needed in feature-based algorithms. It also increase the computation time. Therefore, the proposed algorithm consists of three phases, which are an initial 3D model generation, model evaluation, and model refinement phases, in order to acquire an efficient 3D model. Intensity gradients and incremental Delaunay triangulation are used in the Initial model generation. In this phase, a morphological edge operator is adopted for a fast edge filtering, and the incremental Delaunay triangulation is modified to decrease the computation time by avoiding the calculation errors of whole pixels and selecting a vertex at the near of the centroid within the previous triangle. After the model generation, sparse vertices are matched, then the faces are evaluated with the size, approximation error, and disparity fluctuation of the face in evaluation stage. Thereafter, the faces which have a large error are selectively refined into smaller faces. Experimental results showed that the proposed algorithm could acquire an adaptive model with less modeling errors for both smooth and abrupt areas and could remarkably reduce the model acquisition time.

키워드

참고문헌

  1. Sungye Kim, B. T. Choi, 'Image-based Relighting for the Contents in Virtual Environment,' IPIU'2002, Cheju, Jan, 2002
  2. R. Koch, '3-D Modeling of Human Heads from Stereoscopic Image Sequences,' Conference of the German Pattern Recognition Society 96, Heidelberg, Germany, Sept. 1996
  3. S. Malassiotis and M. G. Strintzis, 'Object-Based Coding of Stereo Image Sequence Using Three-Dimensional Models,' IEEE Trans. on Circuits and Systems for Video Technology, vol. 7, no. 6, pp. 892-905, 1997 https://doi.org/10.1109/76.644069
  4. M. G. Strintzis and S. Malassiotis, 'Object-Based Coding of Stereoscopic and 3D Image Sequences,' IEEE Signal Processing magazine, vol. 16, no. 3, pp. 14-28, 1999 https://doi.org/10.1109/79.768570
  5. D. Tzovaras, N. Grammalidis, and M. G. Strintzis, 'Disparity Field and Depth Map Coding for Multiview 3D Image Generation,' Image Communication, vol. 11, pp. 205-230, 1998 https://doi.org/10.1016/S0923-5965(97)00029-5
  6. R. Lengagne and P. Fua, 'Incorporating Differential Constraints in the 3D Reconstruction Process : Application to Stereo,' In International Conference on Computer Vision, Vancouver, Canada, July 2001, pp. 573-580 https://doi.org/10.1109/ICCV.2001.937569
  7. E. Izquierdo and X. Feng, 'Modeling Arbitrary Objects Based on Geometric Surface Conformity,' IEEE Trans. on Circuits and Systems for Video Technology, vol. 9, no. 2, pp, 336-352, 1999 https://doi.org/10.1109/76.752100
  8. Ho-Keun Lee, Kyu-Phil Han, and Yeong-Ho Ha, 'Hierarchical 3D Modeling Using Disparity-Motion Relationship and Feature Points,' The Journal of The Institute of Electronics Engineers of Korea, vol. 39, no. 1, pp. 9-16, Jan. 2002
  9. C. Ballard and A. Zisserman, 'Automatic Reconstruction of Piecewise Planar Models from Multiple Views,' Computer Vision and Pattern Recognition, vol. 2, pp. 559-565, Jun. 1999 https://doi.org/10.1109/CVPR.1999.784966
  10. Paul E. Debevec, 'Modeling and Rendering Architecture from Photographs,' Ph. D. Thesis, University of California at Berkeley, 1996
  11. J. O'Rourke, Computational Geometry in C 2nd, Cambridge University Press, 1998, pp 161-162
  12. D. F. Watson, 'Computing the n-dimensional Delaunay tessellation with application to Voronoi polytopes,' Comput. J., vol 24, no. 2, pp. 167-172, 1981 https://doi.org/10.1093/comjnl/24.2.167
  13. Robert M. Haralick and Linda G. Shapiro, 'Computer and Robot Vision vol. 1,' pp. 28-48
  14. Rafael C. Gonzalez, Richard E. Woods, 'Digital Image Processing 2nd,' Prentice Hall, 2002. p. 525
  15. http://www.ph.tn.tudelft.nl/~lucas
  16. Takeo Kanade, Masatoshi Okutomi, 'A Stereo Matching Algorithm with an Adaptive Window: Theory and Experiment,' IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 16, no. 9, Sept, 1994 https://doi.org/10.1109/34.310690