3D Orientation Estimation of Industrial Parts from 2D Images using Neural Networks

Julien Langlois, Harold Mouchère, Nicolas Normand, Christian Viard-Gaudin

2018

Abstract

In this paper we propose a pose regression method employing a convolutional neural network (CNN) fed with single 2D images to estimate the 3D orientation of a specific industrial part. The network training dataset is generated by rendering pose-views from a textured CAD model to compensate for the lack of real images and their associated position label. Using several lighting conditions and material reflectances increases the robustness of the prediction and allows to anticipate challenging industrial situations. We show that using a geodesic loss function, the network is able to estimate a rendered view pose with a 5 accuracy while inferring from real images gives visually convincing results suitable for any pose refinement processes.

Paper Citation

in Harvard Style

Langlois J., Mouchère H., Normand N. and Viard-Gaudin C. (2018). 3D Orientation Estimation of Industrial Parts from 2D Images using Neural Networks.In Proceedings of the 7th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM, ISBN 978-989-758-276-9, pages 409-416. DOI: 10.5220/0006597604090416

in Bibtex Style

@conference{icpram18,
author={Julien Langlois and Harold Mouchère and Nicolas Normand and Christian Viard-Gaudin},
title={3D Orientation Estimation of Industrial Parts from 2D Images using Neural Networks},
booktitle={Proceedings of the 7th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,},
year={2018},
pages={409-416},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0006597604090416},
isbn={978-989-758-276-9},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 7th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,
TI - 3D Orientation Estimation of Industrial Parts from 2D Images using Neural Networks
SN - 978-989-758-276-9
AU - Langlois J.
AU - Mouchère H.
AU - Normand N.
AU - Viard-Gaudin C.
PY - 2018
SP - 409
EP - 416
DO - 10.5220/0006597604090416