Area-Based Computation of Stereo Disparity with Model-Based Window Size Selection

Abstract

In direct area-based stereo problems occur if there is a significant local deformation between the two images, and at borders of objects. We propose a correlation method with affine deformable windows in which an optimal window size is selected using model-based reasoning. Two criteria for the window size selection are studied and the resulting algorithms are applied to a random dot pair, an autostereogram and Hoff's fruit pair. The results are found to be both accurate and robust.

Cite

Text

Maas et al. "Area-Based Computation of Stereo Disparity with Model-Based Window Size Selection." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 1999. doi:10.1109/CVPR.1999.786925

Markdown

[Maas et al. "Area-Based Computation of Stereo Disparity with Model-Based Window Size Selection." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 1999.](https://mlanthology.org/cvpr/1999/maas1999cvpr-area/) doi:10.1109/CVPR.1999.786925

BibTeX

@inproceedings{maas1999cvpr-area,
  title     = {{Area-Based Computation of Stereo Disparity with Model-Based Window Size Selection}},
  author    = {Maas, Robert and ter Haar Romeny, Bart M. and Viergever, Max A.},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year      = {1999},
  pages     = {1106-1112},
  doi       = {10.1109/CVPR.1999.786925},
  url       = {https://mlanthology.org/cvpr/1999/maas1999cvpr-area/}
}