ML Anthology

Image Registration Accuracy Estimation Without Ground Truth Using Bootstrap

Jan Kybic, Daniel Smutek
ECCV 2006 pp. 61-72
doi:10.1007/11889762_6 /eccv/2006/kybic2006eccv-image/

Abstract

We consider the problem of estimating the local accuracy of image registration when no ground truth data is available. The technique is based on a statistical resampling technique called bootstrap. Only the two input images are used, no other data are needed. The general bootstrap uncertainty estimation framework described here is in principle applicable to most of the existing pixel based registration techniques. In practice, a large computing power is required. We present experimental results for a block matching method on an ultrasound image sequence for elastography with both known and unknown deformation field.

PDF ECCV Semantic Scholar

Cite

Text

Kybic and Smutek. "Image Registration Accuracy Estimation Without Ground Truth Using Bootstrap." European Conference on Computer Vision, 2006. doi:10.1007/11889762_6

Markdown

[Kybic and Smutek. "Image Registration Accuracy Estimation Without Ground Truth Using Bootstrap." European Conference on Computer Vision, 2006.](https://mlanthology.org/eccv/2006/kybic2006eccv-image/) doi:10.1007/11889762_6

BibTeX

@inproceedings{kybic2006eccv-image,
  title     = {{Image Registration Accuracy Estimation Without Ground Truth Using Bootstrap}},
  author    = {Kybic, Jan and Smutek, Daniel},
  booktitle = {European Conference on Computer Vision},
  year      = {2006},
  pages     = {61-72},
  doi       = {10.1007/11889762_6},
  url       = {https://mlanthology.org/eccv/2006/kybic2006eccv-image/}
}