Abstract
Many novel local image descriptors (Random Ferns, ORB etc) are being proposed each year with claims of being as good as or superior to SIFT for representing point features. In this context we design a simple experimental framework to compare the performances of different descriptors for realtime recognition of 3D points in a given environment. We use this framework to show that robust descriptors like SIFT perform far better when compared to fast binary descriptors like ORB if matching process uses approximate nearest-neighbor search (ANNS) for acceleration. Such an analysis can be very useful for making appropriate choice from vast number of descriptors available in the literature. We further apply machine learning techniques to obtain better approximation of SIFT descriptor matching than ANNS. Though we could not improve its performance, our in-depth analysis of its root cause provides useful insights for guiding future exploration in this topic.
Chapter PDF
Similar content being viewed by others
References
Affine Covariant Regions data from Visual Geometry Group, Oxford University. http://www.robots.ox.ac.uk/~vgg/research/affine/
Matlab functions for multiple view geometry. http://www.robots.ox.ac.uk/~vgg/hzbook/code/
Bay, H., Ess, A., Tuytelaars, T., Van Gool, L.: Speeded-up robust features (SURF). Computer Vision and Image Understanding 110(3), 346–359 (2008)
Bhat. S., Berger, M.O., Sur, F.: Visual words for 3D reconstruction and pose computation. 3DIMPVT 2011 (2011)
Bradski, G.: The OpenCV Library. Dr. Dobb’s Journal of Software Tools (2000)
Calonder, M., Lepetit, V., Ozuysal, M., Trzcinski, T., Strecha, C., Fua, P.: BRIEF: Computing a Local Binary Descriptor Very Fast. PAMI 2012 (2012)
Criminisi, A., Shotton, J.: Decision Forests for Computer Vision and Medical Image Analysis. Springer Publishing Company, Incorporated (2013)
Dahl, A.L., Aanæs, H., Pedersen, K.S.: Finding the best feature detector-descriptor combination. In: 3DIMPVT 2011 (2011)
Donoser, M., Schmalstieg, D.: Discriminative feature-to-point matching in image-based localization. In: CVPR 2014 (2014)
Fischler, M.A., Bolles, R.C.: Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography. Communications of the ACM 24(6), 381–395 (1981)
Hartmann, J., Klussendorff, J., Maehle, E.: A comparison of feature descriptors for visual slam. In: European Conference on Mobile Robots 2013 (2013)
Heinly, J., Dunn, E., Frahm, J.-M.: Comparative evaluation of binary features. In: Fitzgibbon, A., Lazebnik, S., Perona, P., Sato, Y., Schmid, C. (eds.) Computer Vision-ECCV 2012. LNCS, vol. 2012, pp. 759–773. Springer, Heidelberg (2012)
Hesch, J., Roumeliotis, S.: A direct least-squares (dls) method for pnp. In: ICCV 2011 (2011)
Lepetit, V., Fua, P.: Keypoint Recognition using Randomized Trees. IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI) (2006)
Lepetit, V., Moreno-Noguer, F., Fua, P.: EP\(n\)P: An Accurate \(O(n)\) Solution to the P\(n\)P Problem. IJCV 2009 (2009). http://cvlab.epfl.ch/software/EPnP/
Leutenegger, S., Chli, M., Siegwart, R.: Brisk: Binary robust invariant scalable keypoints. In: ICCV 2011 (2011)
Lieberknecht, S., Benhimane, S., Meier, P., Navab, N.: A dataset and evaluation methodology for template-based tracking algorithms. In: ISMAR (2009)
Lowe, D.G.: Distinctive image features from scale-invariant keypoints. IJCV 2004 (2004)
Mikolajczyk, K., Schmid, C.: A performance evaluation of local descriptors. IEEE Transactions on Pattern Analysis and Machine Intelligence 27(10), 1615–1630 (2005)
Moosmann, F., Nowak, E., Jurie, F.: Randomized clustering forests for image classification. PAMI 2008 (2008)
Mount, D.M., Arya, S.: ANN: A library for approximate nearest neighbor searching. http://www.cs.umd.edu/~mount/ANN
Ozuysal, M., Calonder, M., Lepetit, V., Fua, P.: Fast Keypoint Recognition using Random Ferns. PAMI 2012 (2012)
Rublee, E., Rabaud, V., Konolige, K., Bradski, G.: Orb: An efficient alternative to sift or surf. ICCV 2011 (2011)
Shotton, J., Glocker, B., Zach, C., Izadi, S., Criminisi, A., Fitzgibbon, A.: Scene coordinate regression forests for camera relocalization in rgb-d images. CVPR 2013 (2013)
Snavely, N., Seitz, S.M., Szeliski, R.: Photo tourism: exploring photo collections in 3d. ACM Trans. Graph. 2006 (2006)
Sun, L., Ji, S., Ye, J.: Canonical correlation analysis for multilabel classification: A least-squares formulation, extensions, and analysis. PAMI 2011 (2011). http://www.public.asu.edu/~jye02/Software/CCA/index.html
Trzcinski, V.L.T., Christoudias, M., Fua, P.: Boosting Binary Keypoint Descriptors. Computer Vision and Pattern Recognition 2013 (2013)
Wu, C.: Towards linear-time incremental structure from motion. In: 3DV 2013 (2013)
Zheng, Y., Kuang, Y., Sugimoto, S., Astrom, K., Okutomi, M.: Revisiting the pnp problem: a fast, general and optimal solution. In: ICCV 2013 (2013)
Zhou, Q.-Y., Koltun, V.: Dense scene reconstruction with points of interest. ACM Trans. Graph. 2013 (2013). http://www.stanford.edu/~qianyizh/projects/scenedata.html
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer International Publishing Switzerland
About this paper
Cite this paper
Bhat, K.K.S., Kannala, J., Heikkilä, J. (2015). 3D Point Representation For Pose Estimation: Accelerated SIFT vs ORB. In: Paulsen, R., Pedersen, K. (eds) Image Analysis. SCIA 2015. Lecture Notes in Computer Science(), vol 9127. Springer, Cham. https://doi.org/10.1007/978-3-319-19665-7_7
Download citation
DOI: https://doi.org/10.1007/978-3-319-19665-7_7
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-19664-0
Online ISBN: 978-3-319-19665-7
eBook Packages: Computer ScienceComputer Science (R0)