MC-DCNN: Dilated Convolutional Neural Network for Computing Stereo Matching Cost

Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10636)


Designing a model for computing better matching cost is a fundamental problem in stereo method. In this paper, we propose a novel convolutional neural network (CNN) architecture, which is called MC-DCNN, for computing matching cost of two image patches. By adding dilated convolution, our model gains a larger receptive field without adding parameters and losing resolution. We also concatenate the features of last three convolutional layers as a better descriptor that contains information of different image levels. The experimental results on Middlebury datasets validate that the proposed method outperforms the baseline CNN network on stereo matching problem, and especially performs well on weakly-textured areas, which is a shortcoming of traditional methods.


Stereo method Matching cost CNN 


  1. 1.
    Scharstein, D., Szeliski, R., Zabih, R.: A taxonomy and evaluation of dense two-frame stereo correspondence algorithms. In: IEEE SMBV, pp. 131–140 (2001)Google Scholar
  2. 2.
    Hirschmuller, H.: Stereo processing by semiglobal matching and mutual information. IEEE Trans. PAMI 30(2), 328–341 (2008)CrossRefGoogle Scholar
  3. 3.
    Woodford, O., Torr, P., Reid, I.: Global stereo reconstruction under second-order smoothness priors. IEEE Trans. PAMI 31(12), 2115–2128 (2009)CrossRefGoogle Scholar
  4. 4.
    Zbontar, J., LeCun, Y.: Computing the stereo matching cost with a convolutional neural network. In: IEEE CVPR, pp. 1592–1599 (2015)Google Scholar
  5. 5.
    Zbontar, J., LeCun, Y.: Stereo matching by training a convolutional neural network to compare image patches. JMLR 17(1), 2287–2318 (2016)MATHGoogle Scholar
  6. 6.
    Birchfield, S., Tomasi, C.: Depth discontinuities by pixel-to-pixel stereo. IJCV 35(3), 269–293 (1999)CrossRefGoogle Scholar
  7. 7.
    Kong, D., Tao, H.: A method for learning matching errors for stereo computation. BMVC 1, 2–11 (2004)Google Scholar
  8. 8.
    Heo, Y.S., Lee, K.M., Lee, S.U.: Robust stereo matching using adaptive normalized cross-correlation. IEEE Trans. PAMI 33(4), 807–822 (2011)CrossRefGoogle Scholar
  9. 9.
    Hirschmuller, H., Innocent, P.R., Garibaldi, J.: Real-time correlation-based stereo vision with reduced border errors. IJCV 47(1–3), 229–246 (2002)CrossRefMATHGoogle Scholar
  10. 10.
    Hirschmuller, H., Scharstein, D.: Evaluation of stereo matching costs on images with radiometric differences. IEEE Trans. PAMI 31(9), 1582–1599 (2009)CrossRefGoogle Scholar
  11. 11.
    Scharstein, D., Hirschmüller, H., Kitajima, Y., Krathwohl, G., Nešić, N., Wang, X., Westling, P.: High-resolution stereo datasets with subpixel-accurate ground truth. In: Jiang, X., Hornegger, J., Koch, R. (eds.) GCPR 2014. LNCS, vol. 8753, pp. 31–42. Springer, Cham (2014). doi: 10.1007/978-3-319-11752-2_3 Google Scholar
  12. 12.
    Chen, L.C., Papandreou, G., Kokkinos, I.: Semantic image segmentation with deep convolutional nets and fully connected CRFs. arXiv preprint arXiv:1412.7062 (2014)
  13. 13.
    Yu, F., Koltun, V.: Multi-scale context aggregation by dilated convolutions. arXiv preprint arXiv:1511.07122 (2015)
  14. 14.
    Zhang, K., Lu, J., Lafruit, G.: Cross-based local stereo matching using orthogonal integral images. IEEE Trans. CSVT 19(7), 1073–1079 (2009)Google Scholar
  15. 15.
    Trzcinski, T., Christoudias, M., Lepetit, V.: Learning image descriptors with boosting. IEEE Trans. PAMI 37(3), 597–610 (2013)CrossRefGoogle Scholar
  16. 16.
    Simonyan, K., Vedaldi, A., Zisserman, A.: Learning local feature descriptors using convex optimisation. IEEE Trans. PAMI 36(8), 1573–1585 (2014)CrossRefGoogle Scholar
  17. 17.
    Zagoruyko, S., Komodakis, N.: Learning to compare image patches via convolutional neural networks. In: IEEE CVPR, pp. 4353–4361 (2015)Google Scholar
  18. 18.
    Barron, J.T., Poole, B.: The fast bilateral solver. In: ECCV, pp. 617–632 (2016)Google Scholar
  19. 19.
    Kim, K.R., Kim, C.S.: Adaptive smoothness constraints for efficient stereo matching using texture and edge information. In: IEEE ICIP, pp. 3429–3433 (2016)Google Scholar
  20. 20.
    Drouyer, S., Beucher, S., Bilodeau, M., Moreaud, M.: Sparse stereo disparity map densification using hierarchical image segmentation. In: ISMM, pp. 172–184 (2017)Google Scholar
  21. 21.
    Luo, W., Schwing, A.G., Urtasun, R.: Efficient deep learning for stereo matching. In: IEEE CVPR, pp. 5695–5703 (2016)Google Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  1. 1.School of Software EngineeringTongji UniversityShanghaiChina
  2. 2.College of Architecture and Urban PlanningTongji UniversityShanghaiChina

Personalised recommendations