Real-time Detection of 2D Tool Landmarks with Synthetic Training Data

Bram Vanherle, Jeroen Put, Nick Michiels, Frank Van Reeth

2021

Abstract

In this paper a deep learning architecture is presented that can, in real time, detect the 2D locations of certain landmarks of physical tools, such as a hammer or screwdriver. To avoid the labor of manual labeling, the network is trained on synthetically generated data. Training computer vision models on computer generated images, while still achieving good accuracy on real images, is a challenge due to the difference in domain. The proposed method uses an advanced rendering method in combination with transfer learning and an intermediate supervision architecture to address this problem. It is shown that the model presented in this paper, named Intermediate Heatmap Model (IHM), generalizes to real images when trained on synthetic data. To avoid the need for an exact textured 3D model of the tool in question, it is shown that the model will generalize to an unseen tool when trained on a set of different 3D models of the same type of tool. IHM is compared to two existing approaches to keypoint detection and it is shown that it outperforms those at detecting tool landmarks, trained on synthetic data.

Paper Citation

in Harvard Style

Vanherle B., Put J., Michiels N. and Van Reeth F. (2021). Real-time Detection of 2D Tool Landmarks with Synthetic Training Data. In Proceedings of the 2nd International Conference on Robotics, Computer Vision and Intelligent Systems - Volume 1: ROBOVIS, ISBN 978-989-758-537-1, pages 40-47. DOI: 10.5220/0010689900003061

in Bibtex Style

@conference{robovis21,
author={Bram Vanherle and Jeroen Put and Nick Michiels and Frank Van Reeth},
title={Real-time Detection of 2D Tool Landmarks with Synthetic Training Data},
booktitle={Proceedings of the 2nd International Conference on Robotics, Computer Vision and Intelligent Systems - Volume 1: ROBOVIS,},
year={2021},
pages={40-47},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010689900003061},
isbn={978-989-758-537-1},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 2nd International Conference on Robotics, Computer Vision and Intelligent Systems - Volume 1: ROBOVIS,
TI - Real-time Detection of 2D Tool Landmarks with Synthetic Training Data
SN - 978-989-758-537-1
AU - Vanherle B.
AU - Put J.
AU - Michiels N.
AU - Van Reeth F.
PY - 2021
SP - 40
EP - 47
DO - 10.5220/0010689900003061