ValidNet: A Deep Learning Network for Validation of Surface Registration

Joy Mazumder, Joy Mazumder, Mohsen Zand, Mohsen Zand, Sheikh Ziauddin, Sheikh Ziauddin, Michael Greenspan, Michael Greenspan, Michael Greenspan

2020

Abstract

This paper proposes a novel deep learning architecture called ValidNet to automatically validate 3D surface registration algorithms for object recognition and pose determination tasks. The performance of many tasks such as object detection mainly depends on the applied registration algorithms, which themselves are susceptible to local minima. Revealing this tendency and verifying the success of registration algorithms is a difficult task. We treat this as a classification problem, and propose a two-class classifier to distinguish clearly between true positive and false positive instances. Our proposed ValidNet deploys a shared mlp architecture which works on the raw and unordered numeric data of scene and model points. This network is able to perform two fundamental tasks of feature extraction and similarity matching using the powerful capability of deep neural network. Experiments on a large synthetic datasets show that the proposed method can effectively be used in automatic validation of registration.

Paper Citation

in Harvard Style

Mazumder J., Zand M., Ziauddin S. and Greenspan M. (2020). ValidNet: A Deep Learning Network for Validation of Surface Registration. In Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2020) - Volume 4: VISAPP; ISBN 978-989-758-402-2, SciTePress, pages 389-397. DOI: 10.5220/0008988103890397

in Bibtex Style

@conference{visapp20,
author={Joy Mazumder and Mohsen Zand and Sheikh Ziauddin and Michael Greenspan},
title={ValidNet: A Deep Learning Network for Validation of Surface Registration},
booktitle={Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2020) - Volume 4: VISAPP},
year={2020},
pages={389-397},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0008988103890397},
isbn={978-989-758-402-2},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2020) - Volume 4: VISAPP
TI - ValidNet: A Deep Learning Network for Validation of Surface Registration
SN - 978-989-758-402-2
AU - Mazumder J.
AU - Zand M.
AU - Ziauddin S.
AU - Greenspan M.
PY - 2020
SP - 389
EP - 397
DO - 10.5220/0008988103890397
PB - SciTePress