Peter Hönig

I am a research assistant in the Vision for Robotics (V4R) group of Markus Vincze at the Automation and Control Institute at TU Wien.

I focus on improving synthetic dataset generation, object detection and pose estimation for robot vision. The goal of my research is to make robot vision more reliable for object grasping, especially for challenging material properties such as metallic, textureless or transparent objects and ideally without the need for depth information.

selected publications

Improving 2D-3D Dense Correspondences with Diffusion Models for 6D Object Pose Estimation

Peter Hönig , Stefan Thalhammer , and Markus Vincze

AIRoV – The First Austrian Symposium on AI, Robotics, and Vision, 2024

Bib HTML

@article{honig2024improving,
  title = {Improving 2D-3D Dense Correspondences with Diffusion Models for 6D Object Pose Estimation},
  author = {H{\"o}nig, Peter and Thalhammer, Stefan and Vincze, Markus},
  journal = {AIRoV – The First Austrian Symposium on AI, Robotics, and Vision},
  year = {2024},
}

STAR: Shape-focused Texture Agnostic Representations for Improved Object Detection and 6D Pose Estimation

Peter Hönig , Stefan Thalhammer , Jean-Baptiste Weibel , and 2 more authors

arXiv preprint arXiv:2402.04878, 2024

Bib HTML

@article{honig2024star,
  title = {STAR: Shape-focused Texture Agnostic Representations for Improved Object Detection and 6D Pose Estimation},
  author = {H{\"o}nig, Peter and Thalhammer, Stefan and Weibel, Jean-Baptiste and Hirschmanner, Matthias and Vincze, Markus},
  journal = {arXiv preprint arXiv:2402.04878},
  year = {2024},
}