Few-shot Learning of Part-specific Probability Space for 3D Shape Segmentation

Introduction

This is the source code for our paper Few-shot Learning of Part-specific Probability Space for 3D Shape Segmentation

Network Architecture

The architecture of our proposed model is as follows

---

Illustration of shape morphing.

Installation

• Our code is tested on Tensorflow>1.4 and CUDA 8.0

• Clone this repo

git clone https://github.com/Lingjing324/Few-Shot-Learning-of-Part-Specific-Probability-Space-for-3D-Shape-Segmentation.git

Dataset

• Download the ShapeNet part dataset from https://www.shapenet.org/ run notebook file data_preprocessing.ipynb to prepare the tfrecords. or: https://github.com/charlesq34/pointnet/blob/master/part_seg/download_data.sh

Training & Test

run the notebook: train_eval.ipynb

Citation

If you find this useful in your research, please consider citing:

@inproceedings{wang2020few,
  title={Few-Shot Learning of Part-Specific Probability Space for 3D Shape Segmentation},
  author={Wang, Lingjing and Li, Xiang and Fang, Yi},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  pages={4504--4513},
  year={2020}
}