DOI QR코드

DOI QR Code

임베디드 GPU에서의 병렬처리를 이용한 모바일 기기에서의 다중뷰 스테레오 정합

Multiview Stereo Matching on Mobile Devices Using Parallel Processing on Embedded GPU

  • 전윤배 (인하대학교 정보통신공학과) ;
  • 박인규 (인하대학교 정보통신공학과)
  • Jeon, Yun Bae (Inha University, Department of Information and Communication Engineering) ;
  • Park, In Kyu (Inha University, Department of Information and Communication Engineering)
  • 투고 : 2019.10.07
  • 심사 : 2019.11.21
  • 발행 : 2019.11.30

초록

다중뷰 스테레오 정합 알고리즘은 시점이 다른 복수의 2차원 영상으로부터 3차원 형상을 복원하기 위해 사용된다. 기존의 다중뷰 스테레오 정합 알고리즘은 단계별로 많은 계산량을 포함하는 복잡한 구조 때문에 고성능 하드웨어에서만 주로 구현되어왔다. 그러나 최근에 모바일 그래픽 프로세서가 발전하면서 충분한 부동소수점 계산 성능이 확보됨에 따라 기존의 PC 환경에서만 수행되었던 복잡한 컴퓨터 비전 알고리즘들이 모바일 GPU에서 구현되고 있다. 본 논문에서는 임베디드 보드의 모바일 GPU에서의 병렬처리를 기반으로 다중뷰 스테레오 알고리즘의 병렬처리를 구현하고 자원이 제한적인 하드웨어에서의 성능 최적화 기법을 제안한다.

Multiview stereo matching algorithm is used to reconstruct 3D shape from a set of 2D images. Conventional multiview stereo algorithms have been implemented on high-performance hardware due to the heavy complexity that contains a large number of calculations in each step. However, as the performance of mobile graphics processors has recently increased rapidly, complex computer vision algorithms can now be implemented on mobile devices like a smartphone and an embedded board. In this paper we parallelize an multiview stereo algorithm using OpenCL on mobile GPU and provide various optimization techniques on the embedded hardware with limited resource.

키워드

참고문헌

  1. S. M. Seitz, et al., "A comparison and evaluation of multiview stereo reconstruction algorithms," In Proc. of IEEE Conference on Computer Vision and Pattern Recognition, vol. 1, pp. 519-528, 2006.
  2. J. Y. Chang, et al., "GPU-friendly multiview stereo reconstruction using surfel representation and graph cuts," Computer Vision and Image Understanding, vol. 115, no. 5, pp. 620-634, 2011. https://doi.org/10.1016/j.cviu.2010.11.017
  3. D. Gallup, et al., "Real-time plane-sweeping stereo with multiple sweeping directions," In Proc. of IEEE Conference on Computer Vision and Pattern Recognition, pp. 1-8, 2007.
  4. A. Munshi, et al., OpenCL programming guide, Pearson Education, 2011.
  5. J. L. Schonberger, E. Zheng, J. M. Frahm, and M. Pollefeys, "Pixelwise view selection for unstructured multiview stereo," In Proc. of European Conference on Computer Vision, 2016.
  6. I. K. Park, et al., "Design and performance evaluation of image proc-essing algorithms on GPUs," IEEE Trans. on Parallel and Distributed Systems, vol. 22, no. 1, pp. 91-104, 2010. https://doi.org/10.1109/TPDS.2010.115
  7. J. E. Stone, D. Gohara, and G. Shi, "OpenCL: A parallel programming standard for heterogeneous computing systems," Computing in Science & Engineering, vol. 12, no. 3, pp. 66, 2010.
  8. G. Wang, et al., "Accelerating computer vision algorithms using OpenCL framework on the mobile GPU-a case study," In Proc. of IEEE International Conference on Acoustics, Speech and Signal Processing, pp. 2629-2633, 2013.
  9. Y. Furukawa and C. Hernandez, "Multiview stereo: A tutorial," Foundations and Trends(R) in Computer Graphics and Vision, vol. 9, no. 1-2, pp. 1-148, 2015. https://doi.org/10.1561/0600000052
  10. A. Ladikos, S. Benhimane, and N. Navab, "Efficient visual hull computation for real-time 3D reconstruction using CUDA," In Proc. of IEEE Conference on Computer Vision and Pattern Recognition Workshops, pp. 1-8, 2008.
  11. N. Singhal, I. K. Park, and S. Cho, "Implementation and optimization of image processing algorithms on hand-held GPU," In Proc. of IEEE International Conference on Image Processing, pp. 4481-4484, 2010.
  12. The Middlebury Computer Vision Pages, http://vision.middlebury.edu/mview/.
  13. C. H. Esteban and F. Schmitt, "Silhouette and stereo fusion for 3D object modeling," Computer Vision and Image Understanding, vol. 96, no. 3, pp. 367-392, 2004. https://doi.org/10.1016/j.cviu.2004.03.016