ParametricNet is a new 6DoF pose estimation network for parametric shapes with keypoint learning and Hough voting scheme.
This is the code of tensorflow version for our ICRA2021 paper: ParametricNet: 6DoF Pose Estimation Network for Parametric Shapes in Stacked Scenarios.
Ubuntu 18.04/20.04
python3.6, tensorflow1.5, opencv-python, sklearn, h5py, et al.
Our backbone PointSIFT is borrowed from pointSIFT.
Siléane dataset is available at here.
Parametric dataset is available at here.
The python code of evaluation metric is available at here.
If you use this codebase in your research, please cite:
@inproceedings{zeng2021parametricnet,
title={ParametricNet: 6DoF Pose Estimation Network for Parametric Shapes in Stacked Scenarios},
author={Zeng, Long and Lv, Wei Jie and Zhang, Xin Yu and Liu, Yong Jin},
booktitle={2021 IEEE International Conference on Robotics and Automation (ICRA)},
pages={772--778},
year={2021},
organization={IEEE}
}