SPU-IMR: Self-supervised Arbitrary-scale Point Cloud Upsampling via Iterative Mask-recovery Network

Environment

Pytorch >= 1.9.0
cuda >= 11.1

Training

python train.py

Dataset

The dataset we used for training and testing is obtained by SAPCU.

https://pan.baidu.com/s/1OPVnCHq129DBMWh5BA2Whg 
access code: hgii

or

https://1drv.ms/f/s!AsP2NtMX-kUTml4U3DYUD6Hy9FJn?e=8QfJTH

Evaluation

a. Download models

Download the pretrained models from the link and unzip it to ./out/

Link: https://pan.baidu.com/s/1KrbKAwGhI1u6buauT6Xd5g?pwd=zxys
Access code: zxys 

b. Evaluation

cd eval
python evaluation.py

c. Points to note

More code for evaluation can be downloaded from:

https://github.com/pleaseconnectwifi/Meta-PU/tree/master/evaluation_code
https://github.com/jialancong/3D_Processing

Acknowledgement

The code is based on Point-MAE, SAPCU and Point-Transformer, Thanks for their great work. If you use any of this code, please cite these works.