ML Anthology

On the Repeatability of the Local Reference Frame for Partial Shape Matching

Alioscia Petrelli, Luigi Di Stefano
ICCV 2011 pp. 2244-2251
doi:10.1109/ICCV.2011.6126503 /iccv/2011/petrelli2011iccv-repeatability/

Abstract

We investigate on local reference frames (LRF) deployed
\nwith 3D descriptors to achieve invariance to objects’ pose.
\nWe address the task of matching together partial views of
\nsurfaces and propose an experimental study on a large corpus
\nof real data which allows for clearly ranking existing
\nLRF proposals based on their repeatability. Then, drawing
\ninspiration from analysis of the experimental findings,
\nwe formulate a new proposal which, in particular, peculiarly
\nincludes a procedure aimed at estimating a repeatable
\nLRF also at border features, which is very important
\nwhen matching partial views of surfaces. Experiments show
\nthat the new proposal neatly outperforms existing methods
\nin terms of repeatability, is computationally very efficient
\nand provide relevant benefits in practical applications.

PDF ICCV Semantic Scholar

Cite

Text

Petrelli and Di Stefano. "On the Repeatability of the Local Reference Frame for Partial Shape Matching." IEEE/CVF International Conference on Computer Vision, 2011. doi:10.1109/ICCV.2011.6126503

Markdown

[Petrelli and Di Stefano. "On the Repeatability of the Local Reference Frame for Partial Shape Matching." IEEE/CVF International Conference on Computer Vision, 2011.](https://mlanthology.org/iccv/2011/petrelli2011iccv-repeatability/) doi:10.1109/ICCV.2011.6126503

BibTeX

@inproceedings{petrelli2011iccv-repeatability,
  title     = {{On the Repeatability of the Local Reference Frame for Partial Shape Matching}},
  author    = {Petrelli, Alioscia and Di Stefano, Luigi},
  booktitle = {IEEE/CVF International Conference on Computer Vision},
  year      = {2011},
  pages     = {2244-2251},
  doi       = {10.1109/ICCV.2011.6126503},
  url       = {https://mlanthology.org/iccv/2011/petrelli2011iccv-repeatability/}
}