Bayesian Optimization of 3D Feature Parameters for 6D Pose Estimation

Frederik Hagelskjær*, Norbert Krüger, Anders Glent Buch

*Kontaktforfatter for dette arbejde

Publikation: Kapitel i bog/rapport/konference-proceedingKonferencebidrag i proceedingsForskningpeer review

94 Downloads (Pure)

Abstrakt

6D pose estimation using local features has shown state-of-the-art performance for object recognition and pose estimation from 3D data in a number of benchmarks. However, this method requires extensive knowledge and elaborate parameter tuning to obtain optimal performances. In this paper, we propose an optimization method able to determine feature parameters automatically, providing improved point matches to a robust pose estimation algorithm. Using labeled data, our method measures the performance of the current parameter setting using a scoring function based on both true and false positive detections. Combined with a Bayesian optimization strategy, we achieve automatic tuning using few labeled examples. Experiments were performed on two recent RGB-D benchmark datasets. The results show significant improvements by tuning an existing algorithm, with state-of-art performance.
OriginalsprogEngelsk
TitelProceedings of the 14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
RedaktørerAndreas Kerren, Christophe Hurter, Jose Braz
Vol/bind5: VISAPP
ForlagSCITEPRESS Digital Library
Publikationsdato2019
Sider135-142
ISBN (Elektronisk)978-989-758-354-4
DOI
StatusUdgivet - 2019
Begivenhed14th International Conference on Computer Vision Theory and Applications - Prague, Tjekkiet
Varighed: 25. feb. 201927. feb. 2019

Konference

Konference14th International Conference on Computer Vision Theory and Applications
LandTjekkiet
ByPrague
Periode25/02/201927/02/2019

Fingeraftryk Dyk ned i forskningsemnerne om 'Bayesian Optimization of 3D Feature Parameters for 6D Pose Estimation'. Sammen danner de et unikt fingeraftryk.

Citationsformater