Improving 6D Object Pose Estimation of metallic Household and Industry Objects

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Pöllabauer, Thomas, Gasser, Michael, Wirth, Tristan, Berkei, Sarah, Knauthe, Volker, Kuijper, Arjan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913720157863936
author Pöllabauer, Thomas
Gasser, Michael
Wirth, Tristan
Berkei, Sarah
Knauthe, Volker
Kuijper, Arjan
author_facet Pöllabauer, Thomas
Gasser, Michael
Wirth, Tristan
Berkei, Sarah
Knauthe, Volker
Kuijper, Arjan
contents 6D object pose estimation suffers from reduced accuracy when applied to metallic objects. We set out to improve the state-of-the-art by addressing challenges such as reflections and specular highlights in industrial applications. Our novel BOP-compatible dataset, featuring a diverse set of metallic objects (cans, household, and industrial items) under various lighting and background conditions, provides additional geometric and visual cues. We demonstrate that these cues can be effectively leveraged to enhance overall performance. To illustrate the usefulness of the additional features, we improve upon the GDRNPP algorithm by introducing an additional keypoint prediction and material estimator head in order to improve spatial scene understanding. Evaluations on the new dataset show improved accuracy for metallic objects, supporting the hypothesis that additional geometric and visual cues can improve learning.
format Preprint
id arxiv_https___arxiv_org_abs_2503_03655
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Improving 6D Object Pose Estimation of metallic Household and Industry Objects
Pöllabauer, Thomas
Gasser, Michael
Wirth, Tristan
Berkei, Sarah
Knauthe, Volker
Kuijper, Arjan
Computer Vision and Pattern Recognition
Artificial Intelligence
6D object pose estimation suffers from reduced accuracy when applied to metallic objects. We set out to improve the state-of-the-art by addressing challenges such as reflections and specular highlights in industrial applications. Our novel BOP-compatible dataset, featuring a diverse set of metallic objects (cans, household, and industrial items) under various lighting and background conditions, provides additional geometric and visual cues. We demonstrate that these cues can be effectively leveraged to enhance overall performance. To illustrate the usefulness of the additional features, we improve upon the GDRNPP algorithm by introducing an additional keypoint prediction and material estimator head in order to improve spatial scene understanding. Evaluations on the new dataset show improved accuracy for metallic objects, supporting the hypothesis that additional geometric and visual cues can improve learning.
title Improving 6D Object Pose Estimation of metallic Household and Industry Objects
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2503.03655