DOI QR코드

DOI QR Code

Depth Image Upsampling Algorithm Using Selective Weight

선택적 가중치를 이용한 깊이 영상 업샘플링 알고리즘

  • Shin, Soo-Yeon (Department of Electronic Engineering, Chungbuk National University) ;
  • Kim, Dong-Myung (Department of Electronic Engineering, Chungbuk National University) ;
  • Suh, Jae-Won (Department of Electronic Engineering, Chungbuk National University)
  • Received : 2017.03.23
  • Accepted : 2017.04.14
  • Published : 2017.07.31

Abstract

In this paper, we present an upsampling technique for depth map image using selective bilateral weights and a color weight using laplacian function. These techniques prevent color texture copy problem, which problem appears in existing upsamplers uses bilateral weight. First, we construct a high-resolution image using the bicubic interpolation technique. Next, we detect a color texture region using pixel value differences of depth and color image. If an interpolated pixel belongs to the color texture edge region, we calculate weighting values of spatial and depth in $3{\times}3$ neighboring pixels and compute the cost value to determine the boundary pixel value. Otherwise we use color weight instead of depth weight. Finally, the pixel value having minimum cost is determined as the pixel value of the high-resolution depth image. Simulation results show that the proposed algorithm achieves good performance in terns of PSNR comparison and subjective visual quality.

본 논문은 양방향 가중치를 이용하는 기존의 업샘플링 방법들에서 나타난 색상 텍스쳐 복사(color texture copy) 문제를 방지하기 위해 선택적 양방향 가중치와 라플라시안 함수를 이용한 색상 가중치를 제안한다. 제안하는 알고리즘은 먼저 3차 회선 보간법(bicubic interpolation)을 통해 높은 해상도의 깊이영상을 생성한다. 그 후 색상영상과 깊이영상의 주변 화소값 차이를 이용하여 색상 텍스쳐 영역을 추정한다. 만일 보간 된 화소가 색상 텍스쳐 영역에 속한다면 해당화소를 포함하는 $3{\times}3$ 영역의 화소들에 대한 거리정보와 깊이정보의 가중치를 구하고 경계 화소값 결정을 위한 비용계산을 수행한다. 반면에 색상 텍스쳐 영역에 포함되지 않는 화소는 깊이정보 가중치 대신 색상정보 가중치를 구하여 비용계산을 수행한다. 아홉 개의 화소에 대한 비용계산이 끝나면 가장 작은 경계 화소값 결정 비용을 가지는 화소 값을 결과영상의 화소값으로 정한다. 제안하는 알고리즘은 PSNR 및 주관적 화질 비교에서 우수한 성능을 보였다.

Keywords

References

  1. R. Nair, K. Ruhl, and F. Lenzen, "A Survey on Time-of-Flight Stereo Fusion," in Time-of-flight and depth imaging. sensors, algorithms and applications, Springer Berlin Heidelberg., ch. 2, pp. 105-127, 2013.
  2. R. Keys, "Cubic convolution interpolation for digital image processing," IEEE Transactions on Acoustics Speech and Signal Processing, vol. 29, no. 6, pp. 1153-1160, Dec. 1981. https://doi.org/10.1109/TASSP.1981.1163711
  3. M. Kopf, F. Cohen, D. Lischinski, and M. Uyttendaele, "Joint Bilateral Upsampling," ACM Transactions on Graphics, vol. 26, no. 3, Jul. 2007.
  4. Y. Li, L. Zhang, Y. Zhang, H. Xuan and Q. Dai, "Depth Map Super-Resolution Via Iterative Joint Trilateral Upsampling," IEEE Conference on Visual Communications and Image Processing, pp. 386-389, 2014.
  5. Q.Yang, R. Yang, and James Davis, "Spatial-Depth Super Resolution for Range Images," IEEE Conference on Computer Vision and Pattern Recognition, pp.1-8, 2007.
  6. Q. Yang, R. Yang, and K. H. Tan, "Fusion of median and bilateral filtering for range image upsampling," IEEE Transactions on Image Processing, vol. 22, no. 12, pp.4841- 4852, Dec. 2013. https://doi.org/10.1109/TIP.2013.2278917
  7. S. Y. Kim and Y. Ho, "Fast Edge Preserving Depth Image Upsampler," IEEE Transactions on Comsumer Electronics, vol. 58, no. 3, pp. 1176-1190, Aug. 2012. https://doi.org/10.1109/TCE.2012.6414983
  8. G. Petschnigg, M. Agrawala, H. Hoppe, R. Szeliski, M. Cohen, and K. Toyama, "Digital photography with flash and no-flash image pairs," ACM Transactions on Graphics, vol. 23, no. 3, pp.664-672, Aug. 2004. https://doi.org/10.1145/1015706.1015777
  9. D. Scharstein, H. Hirschmüller, and P. Westling, "Highresolution stereo datasets with subpixel-accurate ground truth," German Conference on Pattern Recognition, Munster, vol. 8753, pp.31-42, 2014.