ML Anthology

CenDerNet: Center and Curvature Representations for Render-and-Compare 6d Pose Estimation

Peter De Roovere, Rembert Daems, Jonathan Croenen, Taoufik Bourgana, Joris de Hoog, Francis Wyffels
ECCVW 2022 pp. 97-111
doi:10.1007/978-3-031-25085-9_6 /eccvw/2022/roovere2022eccvw-cendernet/

Abstract

We introduce CenDerNet, a framework for 6D pose estimation from multi-view images based on center and curvature representations. Finding precise poses for reflective, textureless objects is a key challenge for industrial robotics. Our approach consists of three stages: First, a fully convolutional neural network predicts center and curvature heatmaps for each view; Second, center heatmaps are used to detect object instances and find their 3D centers; Third, 6D object poses are estimated using 3D centers and curvature heatmaps. By jointly optimizing poses across views using a render-and-compare approach, our method naturally handles occlusions and object symmetries. We show that CenDerNet outperforms previous methods on two industry-relevant datasets: DIMO and T-LESS .

PDF ECCVW Semantic Scholar

Cite

Text

De Roovere et al. "CenDerNet: Center and Curvature Representations for Render-and-Compare 6d Pose Estimation." European Conference on Computer Vision Workshops, 2022. doi:10.1007/978-3-031-25085-9_6

Markdown

[De Roovere et al. "CenDerNet: Center and Curvature Representations for Render-and-Compare 6d Pose Estimation." European Conference on Computer Vision Workshops, 2022.](https://mlanthology.org/eccvw/2022/roovere2022eccvw-cendernet/) doi:10.1007/978-3-031-25085-9_6

BibTeX

@inproceedings{roovere2022eccvw-cendernet,
  title     = {{CenDerNet: Center and Curvature Representations for Render-and-Compare 6d Pose Estimation}},
  author    = {De Roovere, Peter and Daems, Rembert and Croenen, Jonathan and Bourgana, Taoufik and de Hoog, Joris and Wyffels, Francis},
  booktitle = {European Conference on Computer Vision Workshops},
  year      = {2022},
  pages     = {97-111},
  doi       = {10.1007/978-3-031-25085-9_6},
  url       = {https://mlanthology.org/eccvw/2022/roovere2022eccvw-cendernet/}
}