CVPR2020 PF-Net: Point Fractal Network for 3D Point Cloud Completion
This repository is still under constructions.
If you have any questions about the code, please email me. Thanks!
This is the Pytorch implement of CVPR2020 PF-Net: Point Fractal Network for 3D Point Cloud Completion.
Pytorch 1.0.1 Python 3.7.4
cd dataset bash download_shapenet_part16_catagories.sh You can also download the dataset from 链接：https://pan.baidu.com/s/1MavAO_GHa0a6BZh4Oaogug 提取码：3hoe
Change ‘croppointnum’ to control the number of missing points. Change ‘pointscaleslist ’to control different input resolutions. Change ‘D_choose’to control without using D-net.
Show the completion results, the program will generate txt files in 'test-examples'.
python show_CD.pyShow the Chamfer Distances and two metrics in our paper.
We provide some incomplete point cloud in file 'testone'. Use the following code to complete a incomplete point cloud of csv file: ``` python Testcsv.py ``` change ‘infile’and ‘infilereal’to select different incomplete point cloud in ‘testone’
Using Meshlab to visualize the txt files.