Towards Deep Learning-based 6D Bin Pose Estimation in 3D Scan

Lukáš Gajdošech, Lukáš Gajdošech, Viktor Kocur, Viktor Kocur, Martin Stuchlík, Lukáš Hudec, Martin Madaras, Martin Madaras

2022

Abstract

An automated robotic system needs to be as robust as possible and fail-safe in general while having relatively high precision and repeatability. Although deep learning-based methods are becoming research standard on how to approach 3D scan and image processing tasks, the industry standard for processing this data is still analytically-based. Our paper claims that analytical methods are less robust and harder for testing, updating, and maintaining. This paper focuses on a specific task of 6D pose estimation of a bin in 3D scans. Therefore, we present a high-quality dataset composed of synthetic data and real scans captured by a structured-light scanner with precise annotations. Additionally, we propose two different methods for 6D bin pose estimation, an analytical method as the industrial standard and a baseline data-driven method. Both approaches are cross-evaluated, and our experiments show that augmenting the training on real scans with synthetic data improves our proposed data-driven neural model. This position paper is preliminary, as proposed methods are trained and evaluated on a relatively small initial dataset which we plan to extend in the future.

Paper Citation

in Harvard Style

Gajdošech L., Kocur V., Stuchlík M., Hudec L. and Madaras M. (2022). Towards Deep Learning-based 6D Bin Pose Estimation in 3D Scan. In Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2022) - Volume 4: VISAPP; ISBN 978-989-758-555-5, SciTePress, pages 545-552. DOI: 10.5220/0010878200003124

in Bibtex Style

@conference{visapp22,
author={Lukáš Gajdošech and Viktor Kocur and Martin Stuchlík and Lukáš Hudec and Martin Madaras},
title={Towards Deep Learning-based 6D Bin Pose Estimation in 3D Scan},
booktitle={Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2022) - Volume 4: VISAPP},
year={2022},
pages={545-552},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010878200003124},
isbn={978-989-758-555-5},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2022) - Volume 4: VISAPP
TI - Towards Deep Learning-based 6D Bin Pose Estimation in 3D Scan
SN - 978-989-758-555-5
AU - Gajdošech L.
AU - Kocur V.
AU - Stuchlík M.
AU - Hudec L.
AU - Madaras M.
PY - 2022
SP - 545
EP - 552
DO - 10.5220/0010878200003124
PB - SciTePress