Outlier Correction in Image Sequences for the Affine Camera
Abstract
It is widely known that, for the affine camera model, both shape and motion can be factorized directly from the so-called image measurement matrix constructed from image point coordinates. The ability to extract both shape and motion from this matrix by a single SVD operation makes this shape-from-motion approach attractive; however, it can not deal with missing feature points and, in the presence of outliers, a direct SVD to the matrix would yield highly un-reliable shape and motion components. In this paper, we present an outlier correction scheme that iteratively updates the elements of the image measurement matrix. The magni-tude and sign of the update to each element is dependent upon the residual robustly estimated in each iteration. The result is that outliers are corrected and retained, giving im-proved reconstruction and smaller reprojection errors. Our iterative outlier correction scheme has been applied to both synthesized and real video sequences. The results obtained are remarkably good. 1.
Cite
Text
Huynh et al. "Outlier Correction in Image Sequences for the Affine Camera." IEEE/CVF International Conference on Computer Vision, 2003. doi:10.1109/ICCV.2003.1238400Markdown
[Huynh et al. "Outlier Correction in Image Sequences for the Affine Camera." IEEE/CVF International Conference on Computer Vision, 2003.](https://mlanthology.org/iccv/2003/huynh2003iccv-outlier/) doi:10.1109/ICCV.2003.1238400BibTeX
@inproceedings{huynh2003iccv-outlier,
title = {{Outlier Correction in Image Sequences for the Affine Camera}},
author = {Huynh, Du Q. and Hartley, R. and Heyden, Anders},
booktitle = {IEEE/CVF International Conference on Computer Vision},
year = {2003},
pages = {585-590},
doi = {10.1109/ICCV.2003.1238400},
url = {https://mlanthology.org/iccv/2003/huynh2003iccv-outlier/}
}