ImgAdaPoinTr: Improving Point Cloud Completion via Images and Segmentation

Vaagn Chopuryan, Mikhail Kuznetsov, Mikhail Kuznetsov, Vasilii Latonov, Natalia Semenova, Natalia Semenova

2024

Abstract

Point cloud completion is an essential task consisting of inferring and filling in missing parts of a 3D point cloud representation. In this paper, we present an ImgAdaPoinTr model, which extends the original Transformer encoder-decoder architecture by accurately incorporating visual information. Besides, we assumed using segmentation of 3D objects as a part of the pipeline due to acquiring an additional increase in performance. We also introduce the novel ImgPCN dataset generated by our rendering tool. The results show that our approach outperforms AdaPoinTr by average 2.9% and 10.3% in terms of Chamfer-Distance L1 and L2 metrics, respectively. The code and dataset are available via the link https://github.com/ImgAdaPoinTr.

Paper Citation

in Harvard Style

Chopuryan V., Kuznetsov M., Latonov V. and Semenova N. (2024). ImgAdaPoinTr: Improving Point Cloud Completion via Images and Segmentation. In Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP; ISBN 978-989-758-679-8, SciTePress, pages 47-58. DOI: 10.5220/0012398100003660

in Bibtex Style

@conference{visapp24,
author={Vaagn Chopuryan and Mikhail Kuznetsov and Vasilii Latonov and Natalia Semenova},
title={ImgAdaPoinTr: Improving Point Cloud Completion via Images and Segmentation},
booktitle={Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP},
year={2024},
pages={47-58},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012398100003660},
isbn={978-989-758-679-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP
TI - ImgAdaPoinTr: Improving Point Cloud Completion via Images and Segmentation
SN - 978-989-758-679-8
AU - Chopuryan V.
AU - Kuznetsov M.
AU - Latonov V.
AU - Semenova N.
PY - 2024
SP - 47
EP - 58
DO - 10.5220/0012398100003660
PB - SciTePress