ML Anthology

Evaluation of Features Detectors and Descriptors Based on 3D Objects

Pierre Moreels, Pietro Perona
ICCV 2005 pp. 800-807
doi:10.1109/ICCV.2005.89 /iccv/2005/moreels2005iccv-evaluation/

Abstract

We explore the performance of a number of popular feature detectors and descriptors in matching 3D object features across viewpoints and lighting conditions. To this end we design a method, based on intersecting epipolar constraints, for providing ground truth correspondence automatically. We collect a database of 100 objects viewed from 144 calibrated viewpoints under three different lighting conditions. We find that the combination of Hessian-affine feature finder and SIFT features is most robust to viewpoint change. Harris-affine combined with SIFT and Hessian-affine combined with shape context descriptors were best respectively for lighting changes and scale changes. We also find that no detector-descriptor combination performs well with viewpoint changes of more than 25-30/spl deg/.

ICCV Semantic Scholar

Cite

Text

Moreels and Perona. "Evaluation of Features Detectors and Descriptors Based on 3D Objects." IEEE/CVF International Conference on Computer Vision, 2005. doi:10.1109/ICCV.2005.89

Markdown

[Moreels and Perona. "Evaluation of Features Detectors and Descriptors Based on 3D Objects." IEEE/CVF International Conference on Computer Vision, 2005.](https://mlanthology.org/iccv/2005/moreels2005iccv-evaluation/) doi:10.1109/ICCV.2005.89

BibTeX

@inproceedings{moreels2005iccv-evaluation,
  title     = {{Evaluation of Features Detectors and Descriptors Based on 3D Objects}},
  author    = {Moreels, Pierre and Perona, Pietro},
  booktitle = {IEEE/CVF International Conference on Computer Vision},
  year      = {2005},
  pages     = {800-807},
  doi       = {10.1109/ICCV.2005.89},
  url       = {https://mlanthology.org/iccv/2005/moreels2005iccv-evaluation/}
}