SC^2-PCR: A Second Order Spatial Compatibility for Efficient and Robust Point Cloud Registration (CVPR 2022)

Libtorch implementation of the paper:

SC^2-PCR: A Second Order Spatial Compatibility for Efficient and Robust Point Cloud Registration.

Zhi Chen, Kun Sun, Fan Yang, Wenbing Tao.

Original Pytorch version

Requirements

This work has been test on Ubuntu 20.04 and you also need CUDA-toolkit.

PCL

Libtorch

cxx11 ABI version is required

YAML

Build Options:

(Modify your libtorch path in CMakeList.txt before compilation)
mkdir build
cd build
cmake ..

Then run the file:

./build/visualize

The original point cloud and the aligned point cloud are shown on screen