A Caltech Library Service

Evaluation of Features Detectors and Descriptors based on 3D objects

Moreels, Pierre and Perona, Pietro (2005) Evaluation of Features Detectors and Descriptors based on 3D objects. In: Tenth IEEE International Conference on Computer Vision (ICCV'05). Vol.1. IEEE , Piscataway, NJ, pp. 800-807. ISBN 0-7695-2334-X.

[img] PDF - Published Version
See Usage Policy.


Use this Persistent URL to link to this item:


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°.

Item Type:Book Section
Related URLs:
URLURL TypeDescription ItemJournal Article
Perona, Pietro0000-0002-7583-5809
Additional Information:© 2005 IEEE. The authors are grateful to Timor Kadir, Yan Ke, David Lowe, Jiri Matas and Krystian Mikolajczyk for providing part or all of their detectors and descriptors code.
Record Number:CaltechAUTHORS:20180613-155808251
Persistent URL:
Official Citation:P. Moreels and P. Perona, "Evaluation of features detectors and descriptors based on 3D objects," Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1, 2005, pp. 800-807 Vol. 1. doi: 10.1109/ICCV.2005.89
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:87090
Deposited By: George Porter
Deposited On:14 Jun 2018 17:45
Last Modified:15 Nov 2021 20:45

Repository Staff Only: item control page