Abstract
‘Invariant regions’ are self-adaptive image patches that automatically deform with changing viewpoint as to keep on covering identical physical parts of a scene. Such regions can be extracted directly from a single image. They are then described by a set of invariant features, which makes it relatively easy to match them between views, even under wide baseline conditions. In this contribution, two methods to extract invariant regions are presented. The first one starts from corners and uses the nearby edges, while the second one is purely intensity-based. As a matter of fact, the goal is to build an opportunistic system that exploits several types of invariant regions as it sees fit. This yields more correspondences and a system that can deal with a wider range of images. To increase the robustness of the system, two semi-local constraints on combinations of region correspondences are derived (one geometric, the other photometric). They allow to test the consistency of correspondences and hence to reject falsely matched regions. Experiments on images of real-world scenes taken from substantially different viewpoints demonstrate the feasibility of the approach.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Ballester, C. and Gonzalez, M. 1998. Affine invariant texture segmentation and shape from texture by variational methods. Journal of Mathematical Imaging and Vision, 9:141–171.
Baumberg, A. 2000. Reliable feature matching across widely separated views. In Proc. IEEE Conf. on Computer Vision and Pattern Recognition, pp. 774–781.
Canny, J.F. 1986. A computational approach to edge detection. IEEE Trans. on Pattern Analysis and Machine Intelligence, 8(6):679–698.
Carlsson, S. 2000. Recognizing walking people. In Proc. European Conference on Computer Vision, pp. 472–486.
Dufournaud, Y., Schmid C., and Horaud, R. 2000. Matching image with different resolutions. In Proc. IEEE Conf. on ComputerVision and Pattern Recognition, pp. 612–618.
Fischler, M.A. and Bolles, R.C. 1981. Random sampling consensus—A paradigm for model fitting with applications to image analysis and automated cartography. Commun. Assoc. Comp. Mach, 24(6):381–395.
Funt, B., Barnard, K., and Martin, L. 1998. Is colour constancy good enough? In Proc. European Conference on Computer Vision, pp. 445–459.
Gruen, A.W. 1985. Adaptive least squares correlation: A powerful image matching technique. Journal of Photogrammetry, Remote Sensing and Cartography, 14(3):175–187.
Hall, D., Colin de Verdi`ere, V., and Crowley, L. Object recognition using coloured receptive fields. In Proc. European Conference on Computer Vision, pp.164–177.
Harris, C. J. and Stephens, M. 1983. A combined corner and edge detector. In Proc. Alvey Vision Conf., pp. 147–151.
Koenderink, J.J. and Van Doorn, A. 1987. Representation of local geometry in the visual system. Biological Cybernetics, 55:367–375.
Lindeberg, T. 1998. Feature detection with automatic scale selection. Int. Journal of Computer Vision, 30(2):79–116.
Lindeberg, T. and Garding, J. 1997. Shape-adapted smoothing in estimation of 3D depth cues from affine distortions of local, 2D brightness structure. Image and Vision Computing, 15:415–434.
Lowe, D. 1999. Object recognition from local scale-invariant features. In Proc. Int. Conf. on Computer Vision, pp. 1150–1157.
Mindru, F., Moons, T., and Van Gool, L. 1999. Recognizing color patterns irrespective of viewpoint and illumination. In Proc. IEEE Conf. on Computer Vision and Pattern Recognition, pp. 368–373.
Montesinos, P., Gouet, V., and Pele, D. 2000. Matching color uncalibrated images using differential invariants. Image and Vision Computing, Special Issue BMVC2000, Elsevier Science, 18(9):659–671.
Pritchett, P. and Zisserman, A. 1998. Wide baseline stereo. In Proc. Int. Conf. on Computer Vision, pp. 754–759.
Schaffalitzky, F. and Zisserman, A. 2001.Viewpoint invariant texture matching and wide baseline stereo. In Proc. Int. Conf. on Computer Vision, pp. 636–643.
Schmid, C., Mohr, R., and Bauckhage, C. 1997. Local grey-value invariants for image retrieval. Int. Journal on Pattern Analysis and Machine Intelligence, 19(5):872–877.
Schmid, C. and Mohr, R. 1998. Comparing and evaluating interests points. In Proc. Int. Conf. on Computer Vision, pp. 230–235.
Shi, J. and Tomasi, C. 1994. Good features to track. In Proc. Int. Conf. on Computer Vision and Pattern Recognition, pp. 593–600.
Sinclair, D., Christensen, H., and Rothwell, C. 1995. Using the relation between a plane projectivity and the fundamental matrix. In Proc. Scandinavian Conf. on Image Analysis, pp. 181–188.
Super, B.J. and Klarquist, W.N. 1997. Patch matching and stereopsis in a general stereo viewing geometry. Int. Journal on Pattern Analysis and Machine Intelligence, 19(3):247–253.
Tell, D. and Carlsson, S. 2000. Wide baseline point matching using affine invariants computed from intensity profiles. In Proc. European Conf. on Computer Vision, pp. 814–828.
Tuytelaars, T. and Van Gool, L. 1999. Content-based image retrieval based on local affinely invariant regions. In Proc. Int. Conf. on Visual Information Systems, pp. 493–500.
Tuytelaars, T., Van Gool, L., D'haene, L., and Koch, R. 1999. Matching affinely invariant regions for visual servoing. In Proc. Int. Conf. on Robotics and Automation, pp. 1601–1606.
Tuytelaars, T., Turina, A., and Van Gool, L. 2003. Non-combinatorial detection of regular repetitions under perspective skew. IEEE Trans. on Pattern Analysis and Machine Intelligence, 25(4):418–432.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Tuytelaars, T., Van Gool, L. Matching Widely Separated Views Based on Affine Invariant Regions. International Journal of Computer Vision 59, 61–85 (2004). https://doi.org/10.1023/B:VISI.0000020671.28016.e8
Issue Date:
DOI: https://doi.org/10.1023/B:VISI.0000020671.28016.e8