Stereo Matching with Improved Radiometric Invariant Matching Cost and Disparity Refinement

  • Jinjin Shi
  • Fangfa Fu
  • Yao Wang
  • Weizhe Xu
  • Jinxiang WangEmail author
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9771)


Accurate and real-time stereo correspondence is a pressing need for many computer vision applications. In this paper, an improved radiometric invariant matching cost algorithm is proposed. It effectively combines modified census transform with relative gradients measures. Although it is very simple, comparison results on Middlebury stereo testbed demonstrate that it has much lower error rates than many existing algorithms and is very close to the ANCC algorithm which represents the current state of the art under extreme luminance condition but outperforms the ANCC algorithm greatly when there are small radiometric distortions. In addition, we also develop a disparity refinement method with computational complexity invariant to the disparity range. Experimental results on Middlebury datasets show those artifacts near object boundaries are reduced using the proposed disparity refinement method.


Stereo matching Radiometric invariant Census transform Disparity refinement 



This work was supported by a grant from National Natural Science Foundation of China (NSFC, No. 61504032).


  1. 1.
    Heo, Y.S., Lee, K.M., Lee, S.U.: Robust stereo matching using adaptive normalized cross-correlation. IEEE Trans. Pattern Anal. Mach. Intell. 33, 807–822 (2011)CrossRefGoogle Scholar
  2. 2.
    Kim, S., Ham, B., Kim, B., Sohn, K.: Mahalanobis distance cross-correlation for illumination-invariant stereo matching. IEEE Trans. Circ. Syst. Video Technol. 24(11), 1844–1859 (2014)CrossRefGoogle Scholar
  3. 3.
    Zhou, X., Boulanger, P.: Radiometric invariant stereo matching based on relative gradients. In: 19th 9th IEEE International Conference on Image Processing, pp. 2989–2992. IEEE (2001)Google Scholar
  4. 4.
    Viola, P., Wells, W.M.: Alignment by maximization of mutual information. Int. J. Comput. Vis. 24(2), 137–154 (1997)CrossRefGoogle Scholar
  5. 5.
    Fife, W., Archibald, J.: Improved census transforms for resource-optimized stereo vision. IEEE Trans. Circ. Syst. Video Technol. 23(1), 60–73 (2013)CrossRefGoogle Scholar
  6. 6.
    Sinha, S., Scharstein, D., Szeliski R.: Efficient high-resolution stereo matching using local plane sweeps. In: IEEE Conference on Computer Vision and Pattern Recognition, pp. 1582–1589. IEEE (2014)Google Scholar
  7. 7.
    Bleyer, M., Rother, C., Kohli, P., Scharstein D., Sinha S.: Object stereo-joint stereo matching and object segmentation. In: IEEE Conference on Computer Vision and Pattern Recognition, pp. 3081–3088. IEEE (2011)Google Scholar
  8. 8.
    Gu, Z., Su, X., Liu, Y., Zhang, Q.: Local stereo matching with adaptive support-weight, rank transform and disparity calibration. Pattern Recogn. Lett. 29, 1230–1235 (2008)CrossRefGoogle Scholar
  9. 9.
    Hosni, A., Bleyer, M., Gelautz, M., Rhemann, C.: Local stereo matching using geodesic weights. In: 19th IEEE International Conference on Image Processing, pp. 2093–2096. IEEE (2009)Google Scholar
  10. 10.
    Rhemann, C., Hosni, A., Bleyer, M., Rother, C., Gelautz, M.: Fast cost-volume filtering for visual correspondence and beyond. In: IEEE Conference on Computer Vision and Pattern Recognition, pp. 3017–3024. IEEE (2011)Google Scholar
  11. 11.
    Yang, Q.: A non-local cost aggregation method for stereo matching. In: IEEE Conference on Computer Vision and Pattern Recognition, pp. 1402–1409. IEEE (2012)Google Scholar
  12. 12.
    Mei, X., Sun, X., Zhou, M., Jiao, S., Wang, H., Zhang, X.: On building an accurate stereo matching system on graphics hardware. In: IEEE Conference on Computer Vision and Pattern Recognition Workshop, pp. 467–474. IEEE (2011)Google Scholar
  13. 13.
    Sun, X., Mei, X., Jiao, S., Zhou, M., Wang, H.: Stereo matching with reliable disparity propagation. In: International Conference on 3D Imaging, Modeling, Processing, Visualization, Transmission, pp. 132–139, IEEE (2011)Google Scholar
  14. 14.
    Yang, Q., Ji, P., Li, D., Yao, S., Zhang, M.: Fast stereo matching using adaptive guided filtering. Image Vis. Comput. 32(3), 202–211 (2014)CrossRefGoogle Scholar
  15. 15.
    Huang, X., Cui, G., Zhang, Y.: A fast non-local disparity refinement method for stereo matching. In: IEEE International Conference on Image Processing, pp. 3823–3827. IEEE (2014)Google Scholar
  16. 16.
    Yang, Q.: Local smoothness enforced cost volume regularization for fast stereo correspondence. IEEE Signal Process. Lett. 22(9), 1429–1433 (2015)CrossRefGoogle Scholar
  17. 17.
    Birchfield, S., Tomasi, C.: A pixel dissimilarity measure that is insensitive to image sampling. IEEE Trans. Pattern Anal. Mach. Intell. 20(4), 401–406 (1998)CrossRefGoogle Scholar
  18. 18.
    Xu, L., Au, O.C., Sun, W., Fang, L., Zou, F., Li, J.: Stereo matching with optimal local adaptive radiometric compensation. IEEE Signal Process. Lett. 22(2), 131–135 (2015)CrossRefGoogle Scholar
  19. 19.
    Zabih, R., Woodfill, J.: Non-parametric local transforms for computing visual correspondence. In: Eklundh, J.-O. (ed.) ECCV 1994. LNCS, vol. 801, pp. 151–158. Springer, Heidelberg (1994)Google Scholar
  20. 20.
    Miron, A., Ainouz, S., Rogozan, A., Bensrhair, A.: A robust cost function for stereo matching of road scenes. Pattern Recogn. Lett. 38, 70–77 (2014)CrossRefGoogle Scholar
  21. 21.
    Mouats, T., Aouf, N., Richardson, M.: A novel image representation via local frequency analysis for illumination invariant stereo matching. IEEE Trans. Image Process. 24(9), 2685–2700 (2015)MathSciNetCrossRefGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2016

Authors and Affiliations

  • Jinjin Shi
    • 1
  • Fangfa Fu
    • 1
  • Yao Wang
    • 1
  • Weizhe Xu
    • 1
  • Jinxiang Wang
    • 1
    Email author
  1. 1.Microelectronics CenterHarbin Institute of TechnologyHarbinChina

Personalised recommendations