ML Anthology

Motion and Surface Recovery Using Curvature and Motion Consistency

Gilbert Soucy, Frank P. Ferrie
ECCV 1992 pp. 222-226
doi:10.1007/3-540-55426-2_25 /eccv/1992/soucy1992eccv-motion/

Abstract

In this paper we have sketched out an algorithm for reconstructing a sequence of overlapping range images based on two key constraints: minimizing the local variation of curvature across adjacent views, and minimizing the variation of motion parameters across adjacent surface points. It operates without explicitly computing correspondence and without the invoking a global rigidity assumption. Preliminary results indicate that the resulting surface reconstruction is both robust and accurate.

PDF ECCV Semantic Scholar

Cite

Text

Soucy and Ferrie. "Motion and Surface Recovery Using Curvature and Motion Consistency." European Conference on Computer Vision, 1992. doi:10.1007/3-540-55426-2_25

Markdown

[Soucy and Ferrie. "Motion and Surface Recovery Using Curvature and Motion Consistency." European Conference on Computer Vision, 1992.](https://mlanthology.org/eccv/1992/soucy1992eccv-motion/) doi:10.1007/3-540-55426-2_25

BibTeX

@inproceedings{soucy1992eccv-motion,
  title     = {{Motion and Surface Recovery Using Curvature and Motion Consistency}},
  author    = {Soucy, Gilbert and Ferrie, Frank P.},
  booktitle = {European Conference on Computer Vision},
  year      = {1992},
  pages     = {222-226},
  doi       = {10.1007/3-540-55426-2_25},
  url       = {https://mlanthology.org/eccv/1992/soucy1992eccv-motion/}
}