GPU acceleration of the KAZE image feature extraction algorithm

  • B. Ramkumar
  • Rob Laber
  • Hristo Bojinov
  • Ravi Sadananda HegdeEmail author
Original Research Paper


The recently proposed, KAZE image feature detection and description algorithm (Alcantarilla et al. in Proceedings of the British machine vision conference. LNCS, vol 7577, no 6, pp 13.1–13.11, 2013) offers significantly improved robustness in comparison to conventional algorithms like SIFT (scale-invariant feature transform) and SURF (speeded-up robust features). The improved robustness comes at a significant computational cost, however, limiting its use for many applications. We report a GPU acceleration of the KAZE algorithm that is significantly faster than its CPU counterpart. Unlike previous reports, our acceleration does not resort to binary descriptors and can serve as a drop-in replacement for CPU-KAZE, SIFT, SURF etc. By achieving nearly tenfold speedup (for a 1920 by 1200 sized image, our Compute Unified Device Architecture (CUDA)-C implementation took around 245 ms on a single GPU in comparison to nearly 2400 ms for a 16-threaded CPU version) without degradation in feature extraction performance, our work expands the applicability of the KAZE algorithm. Additionally, the strategies described here could also prove useful for the GPU implementation of other nonlinear scale-space-based image processing algorithms.


Nonlinear scale space Feature detection Feature description GPU acceleration KAZE features 



RB and RSH acknowledge funding support from Innit Inc. (Grant no. CNS/INNIT/EE/P0210/1617/007) and High Performance Computing Lab support from Mr. Sudeep Banerjee.


  1. 1.
    Alcantarilla, P., J. Davison, A., Bartoli, A.: Kaze features. In: Proceedings of the British Machine Vision Conference. LNCS. 7577(6), 13.1–13.11 (2013)Google Scholar
  2. 2.
    Alcantarilla, P., Nuevo, J., Bartoli, A.: Fast explicit diffusion for accelerated features in nonlinear scale spaces. In: Proceedings of the British Machine Vision Conference 2013, 13.1–13.11 (2013)Google Scholar
  3. 3.
    Balntas, V., Lenc, K., Vedaldi, A., Mikolajczyk, K.: Hpatches: a benchmark and evaluation of handcrafted and learned local descriptors (2017). arXiv preprint. arXiv:1704.05939
  4. 4.
    Bay, H., Ess, A., Tuytelaars, T., Van Gool, L.: Speeded-up robust features (SURF). Comput. Vis. Image Underst. 110(3), 346–359 (2008)CrossRefGoogle Scholar
  5. 5.
    Bianco, S., Mazzini, D., Pau, D.P., Schettini, R.: Local detectors and compact descriptors for visual search: a quantitative comparison. Digit. Signal Process. Rev. J. 44(1), 1–13 (2015)Google Scholar
  6. 6.
    Björkman, M., Bergström, N., Kragic, D.: Detecting, segmenting and tracking unknown objects using multi-label MRF inference. Comput. Vis. Image Underst. 118, 111–127 (2014)CrossRefGoogle Scholar
  7. 7.
    Calonder, M., Lepetit, V., Strecha, C., Fua, P.: BRIEF: Binary Robust Independent Elementary Features. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). 6314(4), 778–792 (2010)Google Scholar
  8. 8.
    Camargo, A., Papadopoulou, D., Spyropoulou, Z., Vlachonasios, K., Doonan, J.H., Gay, A.P.: Objective definition of rosette shape variation using a combined computer vision and data mining approach. PLoS ONE 9(5):e96889 (2014)CrossRefGoogle Scholar
  9. 9.
    Che, S., Skadron, K.: BenchFriend: correlating the performance of GPU benchmarks. Int. J. High Perform. Comput. Appl. 28(2), 238–250 (2014)CrossRefGoogle Scholar
  10. 10.
    Chen, B., Zhou, X.H., Zhang, L.W., Wang, J., Zhang, W.Q., Zhang, C.: A new nonlinear diffusion equation model for noisy image segmentation. Adv. Math. Phys. 2016 (2016)Google Scholar
  11. 11.
    Chen, Y., Pock, T.: Trainable nonlinear reaction diffusion: a flexible framework for fast and effective image restoration. IEEE Trans. Pattern Anal. Mach. Intell. 39(6), 1256–1272 (2015)CrossRefGoogle Scholar
  12. 12.
    Dalal, N., Triggs, B.: Histograms of oriented gradients for human detection. In: Proceedings—2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR), pp. 886–893 (2005)Google Scholar
  13. 13.
    Feng, L., Wu, Z., Long, X.: Fast image diffusion for feature detection and description. Int. J. Comput. Theory Eng. 8(1), 58–62 (2016)CrossRefGoogle Scholar
  14. 14.
    Fung, J., Mann, S.: Using graphics devices in reverse: GPU-based image processing and computer vision. In: 2008 IEEE International Conference on Multimedia and Expo (ICME), pp. 9–12 (2008)Google Scholar
  15. 15.
    Gao, X., Li, W., Loomes, M., Wang, L.: A fused deep learning architecture for viewpoint classification of echocardiography. Inf. Fusion 36, 103–113 (2017)CrossRefGoogle Scholar
  16. 16.
    Gauglitz, S., Hollerer, T., Turk, M.: Evaluation of interest point detectors and feature descriptors for visual tracking. Int. J. Comput. Vis. 94(3), 335–360 (2011)CrossRefzbMATHGoogle Scholar
  17. 17.
    Grewenig, S., Weickert, J., Bruhn, A.: From box filtering to fast explicit diffusion. In: Goesele M., Roth S., Kuijper A., Schiele B., Schindler K. (eds) Pattern Recognition. DAGM 2010. Lecture Notes in Computer Science (LNCS). 6376, pp. 533–542 (2010).Google Scholar
  18. 18.
    Harvey, R.W., Bosson, A., Bangham, J.A.: A comparison of linear and non-linear scale-space filters in noise. Signal Process. VIII 1(6), 1777–1781 (1996)Google Scholar
  19. 19.
    Heinly, J., Dunn, E., Frahm, J.M.: Comparative evaluation of binary features. In: 12th European Conference on Computer Vision (ECCV), pp. 759–773 (2012)Google Scholar
  20. 20.
    Hu, W., Hu, R., Xie, N., Ling, H., Maybank, S.: Image classification using multiscale information fusion based on saliency driven nonlinear diffusion filtering. IEEE Trans. Image Process. 23(4), 1513–1526 (2014)MathSciNetCrossRefzbMATHGoogle Scholar
  21. 21.
    Huang, H., Lu, L., Yan, B., Chen, J.: A new scale invariant feature detector and modified SURF descriptor. In: Proceedings—2010 6th International Conference on Natural Computation (ICNC). 7, pp. 3734–3738 (2010).Google Scholar
  22. 22.
    Katzourakis, N.: Generalised solutions for fully nonlinear PDE systems and existence-uniqueness theorems. J. Differ. Equ. 263(1), 641–686 (2017)MathSciNetCrossRefzbMATHGoogle Scholar
  23. 23.
    Lehiani, Y., Preda, M., Maidi, M., Ghorbel, F.: Object identification and tracking for steady registration in mobile augmented reality. In: IEEE 2015 International Conference on Signal and Image Processing Applications (ICSIPA), pp. 54–59 (2016)Google Scholar
  24. 24.
    Leutenegger, S., Chli, M., Siegwart, R.Y.: BRISK: binary robust invariant scalable keypoints. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 2548–2555 (2011)Google Scholar
  25. 25.
    Li, Y., Wang, S., Tian, Q., Ding, X.: A survey of recent advances in visual feature detection. Neurocomputing 149(PB), 736–751 (2015)CrossRefGoogle Scholar
  26. 26.
    Lindeberg, T.: Feature detection with automatic scale selection. Int. J. Comput. Vis. 30(2), 79–116 (1998)CrossRefGoogle Scholar
  27. 27.
    Lowe, D.G.: Distinctive image features from scale-invariant keypoints. Int. J. Comput. Vis. 60(2), 91–110 (2004)CrossRefGoogle Scholar
  28. 28.
    Mikolajczyk, K., Mikolajczyk, K., Schmid, C., Schmid, C.: A performance evaluation of local descriptors. IEEE Trans. Pattern Anal. Mach. Intell. 27(10), 1615–1630 (2005)CrossRefGoogle Scholar
  29. 29.
    Mikolajczyk, K., Tuytelaars, T., Schmid, C., Zisserman, A., Matas, J., Schaffalitzky, F., Kadir, T., Gool, L.V.: A comparison of affine region detectors. Int. J. Comput. Vis. 65(1), 43–72 (2005)CrossRefGoogle Scholar
  30. 30.
    Perona, P., Malik, J.: Scale-space and edge detection using anisotropic diffusion. IEEE Trans. Pattern Anal. Mach. Intell. 12(7), 629–639 (1990)CrossRefGoogle Scholar
  31. 31.
    Pieropan, A., Björkman, M., Bergström, N., Kragic, D.: Feature descriptors for tracking by detection: a benchmark (2016). arXiv preprint. arXiv:1607.06178
  32. 32.
    Rosten, E., Drummond, T.: Machine Learning for High-Speed Corner Detection. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). 3951, 430–443 (2006)Google Scholar
  33. 33.
    Rublee, E., Rabaud, V., Konolige, K., Bradski, G.: ORB: An efficient alternative to SIFT or SURF. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 2564–2571 (2011)Google Scholar
  34. 34.
    Sankaran, B., Ramalingam, S., Taguchi, Y.: Parameter learning for improving binary descriptor matching. In: 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 4892–4897 (2016)Google Scholar
  35. 35.
    Sanna, A., Lamberti, F.: Advances in target detection and tracking in forward-looking infrared (FLIR) imagery. Sensors (Switz.) 14(11), 20297–20303 (2014)CrossRefGoogle Scholar
  36. 36.
    Seung, I.P., Ponce, S.P., Huang, J., Cao, Y., Quek, F.: Low-cost, high-speed computer vision using NVIDIA’s CUDA architecture. In: Proceedingsof the 37th IEEE Applied Imagery Pattern Recognition Workshop, pp. 1-7 (2008)Google Scholar
  37. 37.
    Tombari, F., Di Stefano, L., Mattoccia, S., Galanti, A.: Performance evaluation of robust matching measures. In: International Conference on Computer Vision Theory and Applications (VISAPP). 1, pp. 473–478 (2008)Google Scholar
  38. 38.
    Weickert, J., Romeny, B.M.T.H., Viergever, M.A.: Efficient and reliable schemes for nonlinear diffusion filtering. IEEE Trans. Image Process. 7(3), 398–410 (1998)CrossRefGoogle Scholar
  39. 39.
    Weickert, J., Scharr, H.: A scheme for coherence-enhancing diffusion filtering with optimized rotation invariance. J. Vis. Commun. Image Represent. 13(1–2), 103–118 (2002)CrossRefGoogle Scholar
  40. 40.
    Zhai, Y., Ong, Y.S., Tsang, I.W.: The emerging big dimensionality. IEEE Comput. Intell. Mag. 9(3), 14–26 (2014)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2019

Authors and Affiliations

  • B. Ramkumar
    • 1
  • Rob Laber
    • 2
  • Hristo Bojinov
    • 2
  • Ravi Sadananda Hegde
    • 1
    Email author
  1. 1.Department of Electrical EngineeringIndian Institute of TechnologyGandhinagarIndia
  2. 2.Innit Inc.Redwood CityUSA

Personalised recommendations