Advancecd Deep Learning for Computer Vision

Project for ADL4CV: Shape Completion with Meso-Skeleton Learning by Zhisheng Zheng and Dongyue Lu, supervised by Yinyu Nie.

This code is tested under Python 3.6.3, PyTorch 1.2.0 on Ubuntu 18.04 and 20.04.

check the report and poster in the folder docu.

1. Installation

Simply run the following commands.

First, you need to set up pointnet++ dependencies

cd pointnet2
python setup.py install

Then, install knn_cuda by running the following command

pip install --upgrade https://github.com/unlimblue/KNN_CUDA/releases/download/0.2/KNN_CUDA-0.2-py3-none-any.whl

2. Data Preparation

Download training data: airplane, chair, chair-200skeleton, chair-400skeleton, chair-1200skeleton.

Put the data into folder datas.

3. Train

Simply run the following code

python train_PFnet.py

to run the PFnet-based netwotk, or run

python train_PUnet.py

to run the PUnet-based network, or run

python train_PFnet_only.py
python train_PUnet_only.py

to do the ablation study for skeleton. See python train.py --help for all the training options.

4. Test

Simply run the following code

python test_PFnet.py

or

python test_PUnet.py

or

python test_PFnet_only.py
python test_PUnet_only.py

See python test.py --help for all the testing options.

Reference

PU-Net

PF-Net

Point2Skeleton

[1]Charles R Qi, Hao Su, Kaichun Mo, and Leonidas J Guibas. Pointnet: Deep learning on point sets for 3d classification and segmentation. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 652–660, 2017. 1

[2] Charles R Qi, Li Yi, Hao Su, and Leonidas J Guibas. Pointnet++: Deep hierarchical feature learning on point sets in a metric space. arXiv preprint arXiv:1706.02413, 2017. 1

[3] Angela Dai, Charles Ruizhongtai Qi, and Matthias Nießner. Shape completion using 3d-encoder-predictor cnns and shape synthesis. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 5868–5877,2017. 1

[4] Yinyu Nie, Yiqun Lin, Xiaoguang Han, Shihui Guo, Jian Chang, Shuguang Cui, and Jian Jun Zhang. Skeleton-bridged point completion: From global inference to local adjustment. arXiv preprint arXiv:2010.07428, 2020. 1

[5] Lequan Yu, Xianzhi Li, Chi-Wing Fu, Daniel Cohen-Or, and Pheng-Ann Heng. Pu-net: Point cloud upsampling network. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 2790–2799, 2018. 1, 2

[6] Zitian Huang, Yikuan Yu, Jiawen Xu, Feng Ni, and Xinyi Le. Pf-net: Point fractal network for 3d point cloud completion. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 7662–7670, 2020. 2

[7] Cheng Lin, Changjian Li, Yuan Liu, Nenglun Chen, Yi-King Choi, and Wenping Wang. Point2skeleton: Learning skeletal representations from point clouds. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 4277–4286, 2021. 2

[8] Angel X. Chang, Thomas Funkhouser, Leonidas Guibas, Pat Hanrahan, Qixing Huang, Zimo Li, Silvio Savarese, Manolis Savva, Shuran Song, Hao Su, Jianxiong Xiao, Li Yi, and Fisher Yu. Shapenet: An information-rich 3d model repository, 2015. 2