SurfEmb: Dense and Continuous Correspondence Distributions for Object Pose Estimation with Learnt Surface Embeddings

Rasmus Laurvig Haugaard*, Anders Glent Buch*

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

Abstract

We present an approach to learn dense, continuous 2D-3D correspondence distributions over the surface of objects from data with no prior knowledge of visual ambiguities like symmetry. We also present a new method for 6D pose estimation of rigid objects using the learnt distributions to sample, score and refine pose hypotheses. The correspondence distributions are learnt with a contrastive loss, represented in object-specific latent spaces by an encoder-decoder query model and a small fully connected key model. Our method is unsupervised with respect to visual ambiguities, yet we show that the query- and key models learn to represent accurate multi-modal surface distributions. Our pose estimation method improves the state-of-the-art significantly on the comprehensive BOP Challenge, trained purely on synthetic data, even compared with methods trained on real data. The project site is at surfemb.github.io.

Original languageEnglish
Title of host publication2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Number of pages10
PublisherIEEE
Publication date2022
Pages6739-6748
ISBN (Electronic)978-1-6654-6946-3
DOIs
Publication statusPublished - 2022
Event2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) - New Orleans, United States
Duration: 18. Jun 202224. Jun 2022

Conference

Conference2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Country/TerritoryUnited States
CityNew Orleans
Period18/06/202224/06/2022
SeriesIEEE Conference on Computer Vision and Pattern Recognition. Proceedings
ISSN1063-6919

Bibliographical note

Publisher Copyright:
© 2022 IEEE.

Keywords

  • Machine learning
  • Pose estimation and tracking
  • Representation learning

Fingerprint

Dive into the research topics of 'SurfEmb: Dense and Continuous Correspondence Distributions for Object Pose Estimation with Learnt Surface Embeddings'. Together they form a unique fingerprint.

Cite this