Sewer Defect Classification using Synthetic Point Clouds

Joakim Haurum, Moaaz Allahham, Mathias Lynge, Kasper Henriksen, Ivan Nikolov, Thomas Moeslund

Abstract

Sewer pipes are currently manually inspected by trained inspectors, making the process prone to human errors, which can be potentially critical. There is therefore a great research and industry interest in automating the sewer inspection process. Previous research have been focused on working with 2D image data, similar to how inspections are currently conducted. There is, however, a clear potential for utilizing recent advances within 3D computer vision for this task. In this paper we investigate the feasibility of applying two modern deep learning methods, DGCNN and PointNet, on a new publicly available sewer point cloud dataset. As point cloud data from real sewers is scarce, we investigate using synthetic data to bootstrap the training process. We investigate four data scenarios, and find that training on synthetic data and fine-tune on real data gives the best results, increasing the metrics by 6-10 percentage points for the best model. Data and code is available at https://bitbucket.org/aauvap/sewer3dclassification.

Download


Paper Citation


in Harvard Style

Haurum J., Allahham M., Lynge M., Henriksen K., Nikolov I. and Moeslund T. (2021). Sewer Defect Classification using Synthetic Point Clouds.In Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP, ISBN 978-989-758-488-6, pages 891-900. DOI: 10.5220/0010207908910900


in Bibtex Style

@conference{visapp21,
author={Joakim Haurum and Moaaz Allahham and Mathias Lynge and Kasper Henriksen and Ivan Nikolov and Thomas Moeslund},
title={Sewer Defect Classification using Synthetic Point Clouds},
booktitle={Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP,},
year={2021},
pages={891-900},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010207908910900},
isbn={978-989-758-488-6},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP,
TI - Sewer Defect Classification using Synthetic Point Clouds
SN - 978-989-758-488-6
AU - Haurum J.
AU - Allahham M.
AU - Lynge M.
AU - Henriksen K.
AU - Nikolov I.
AU - Moeslund T.
PY - 2021
SP - 891
EP - 900
DO - 10.5220/0010207908910900