Homography and fundamental matrix estimation from region matches using an affine error metric. (English) Zbl 1361.68293

Bentolila, Jacob; Francos, Joseph M.

J. Math. Imaging Vis. 49, No. 2, 481-491 (2014).

Summary: Matching a pair of affine invariant regions between images results in estimation of the affine transformation between the regions. However, the parameters of the affine transformations are rarely used directly for matching images, mainly due to the lack of an appropriate error metric of the distance between them.
In this paper we derive a novel metric for measuring the distance between affine transformations: Given an image region, we show that minimization of this metric is equivalent to the minimization of the mean squared distance between affine transformations of a point, sampled uniformly on the image region. Moreover, the metric of the distance between affine transformations is equivalent to the \(l_{2}\) norm of a linear transformation of the difference between the six parameters of the affine transformations. We employ the metric for estimating homographies and for estimating the fundamental matrix between images. We show that both homography estimation and fundamental matrix estimation methods, based on the proposed metric, are superior to current linear estimation methods as they provide better accuracy without increasing the computational complexity.

MSC:

68U10	Computing methodologies for image processing
68U05	Computer graphics; computational geometry (digital and algorithmic aspects)

Keywords:

epipolar geometry; affine invariant regions; fundamental matrix; homographies

Software:

SIFT

References:

[1]	Bentolila, J.; Francos, J. M., Affine consistency graphs for image representation and elastic matching (2012)
[2]	Cho, M.; Lee, J.; Lee, K. M., Feature correspondence and deformable object matching via agglomerative correspondence clustering, 1280-1287 (2009)
[3]	Chum, O.; Matas, J.; Stepan, O., Epipolar geometry from three correspondences (2003)
[4]	Faugeras, O.: Three-Dimensional Computer Vision: A Geometric Viewpoint. Artificial Intelligence. MIT Press, New York (1993)
[5]	Ferrari, V., Tuytelaars, T., Van Gool, L.: Simultaneous object recognition and segmentation from single or multiple model views. Int. J. Comput. Vis. 67(2), 159-188 (2006)
[6]	Harris, C.; Stephens, M., A combined corner and edge detection, 147-151 (1988)
[7]	Hartley, R., Zisserman, A.: Multiple View Geometry in Computer Vision, 2nd edn. Cambridge University Press, New York (2003) · Zbl 0956.68149
[8]	Hartley, R. I., In defence of the 8-point algorithm, 1064 (1995), Washington
[9]	Hartley, R.I., Zisserman, A.: Multiple View Geometry in Computer Vision Cambridge University Press, Cambridge (2000). ISBN:0521623049 · Zbl 0956.68149
[10]	Kannala, J.; Brandt, S., Quasi-dense wide baseline matching using match propagation, 1-8 (2007)
[11]	Lowe, D.G.: Distinctive image features from scale-invariant keypoints. Int. J. Comput. Vis. 60, 91-110 (2004)
[12]	Matas, J.: Robust wide-baseline stereo from maximally stable extremal regions. Image Vis. Comput. 22(10), 761-767 (2004)
[13]	Mikolajczyk, K.; Schmid, C., An affine invariant interest point detector, 128-142 (2002), Berlin · Zbl 1034.68642
[14]	Mikolajczyk, K., Schmid, C.: Scale & affine invariant interest point detectors. Int. J. Comput. Vis. 60(1), 63-86 (2004)
[15]	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, 2005 (2005) · Zbl 1192.74364
[16]	Moravec, H.: Obstacle avoidance and navigation in the real world by a seeing robot rover. Tech. report CMU-RI-TR-80-03. Robotics Institute, Carnegie Mellon University. CMU-RI-TR-80-03 (1980)
[17]	Perdoch, M.; Matas, J.; Chum, O., Epipolar geometry from two correspondences, No. 04, ICPR ’06, 215-219 (2006), Washington
[18]	Riggi, F.; Toews, M.; Arbel, T., Fundamental matrix estimation via tip—transfer of invariant parameters, No. 02, 21-24 (2006), Washington
[19]	Szeliski, R.; Torr, P., Geometrically constrained structure from motion: points on planes, 171-186 (1998)
[20]	Tola, E.; Lepetit, V.; Fua, P., A fast local descriptor for dense matching, 1-8 (2008)
[21]	Tola, E., Lepetit, V., Fua, P.: Daisy: an efficient dense descriptor applied to wide-baseline stereo. IEEE Trans. Pattern Anal. Mach. Intell. 32(5), 815-830 (2010)
[22]	Tuytelaars, T.; Gool, L. V., Wide baseline stereo matching based on local, affinely invariant regions, 412-425 (2000)
[23]	Tuytelaars, T., Mikolajczyk, K.: Local invariant feature detectors: a survey. Found. Trends Comput. Graph. Vis. 3(3), 177-280 (2008)