1 October 2010 Robust and accurate fundamental matrix estimation with propagated matches
Author Affiliations +
Optical Engineering, 49(10), 107002 (2010). doi:10.1117/1.3490415
Two perspective images of a single scene taken by uncalibrated cameras are related by a fundamental matrix, which is the key to solving many computer vision problems. We present a new robust and accurate algorithm to estimate the fundamental matrix, which is suitable for wide baseline stereo pairs. The estimation algorithm includes two stages. The first stage is that more affine invariant matching points are determined by a propagation process. In this stage, an affine iterative optimization model is used to accurately detect matching points at the subpixel level. And a new matching cost function is proposed that is more robust to outliers and more effective for images with single texture and repetitive texture. In the second stage, a resampling model based on the posterior probability is presented in order to optimize the fundamental matrix with more accurate matching points. The experiment results show that our algorithm is very robust to outliers, and the fundamental matrix with high accuracy can be estimated.
Yanli Wan, Zhenjiang Miao, Zhen Tang, "Robust and accurate fundamental matrix estimation with propagated matches," Optical Engineering 49(10), 107002 (1 October 2010). https://doi.org/10.1117/1.3490415

Optimization (mathematics)

Affine motion model

Reconstruction algorithms

Optical engineering



Computer vision technology


Lawn-mowing algorithm for noisy gradient vector fields
Proceedings of SPIE (September 23 1999)
Surface reconstruction for 3D remote sensing
Proceedings of SPIE (May 07 2012)
Landmark-based navigation: model extension and refinement
Proceedings of SPIE (April 13 1993)
Non-rigid structure from motion based on movement continuity
Proceedings of SPIE (October 28 2011)
Revisit to omni rig sensors What can be done...
Proceedings of SPIE (September 25 2003)

Back to Top