given source set and destination set with normals
, find the best transformation
such that:
[ 1 ]Fischler, Martin A., and Robert C. Bolles. "Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography." Communications of the ACM 24.6 (1981): 381-395.
[ 2 ] Arun, K. Somani, Thomas S. Huang, and Steven D. Blostein. "Least-squares fitting of two 3-D point sets." IEEE Transactions on pattern analysis and machine intelligence 5 (1987): 698-700.
[ 3 ] Low, Kok-Lim. "Linear least-squares optimization for point-to-plane icp surface registration." Chapel Hill, University of North Carolina 4.10 (2004): 1-3.
- verified on:
- ubuntu18.04 | python 3.6 | cuda-11.0
- ubuntu20.04 | python 3.8 | cuda-11.5
-
cuda >= 11
sudo apt install nvidia-driver-XXX- https://developer.nvidia.com/cuda-downloads
- run the installer (ignore warning)
- install the toolkit (no need for the drivers)
-
cmake >= 3.15
- install:
sudo apt remove --purge cmake; sudo snap install cmake --classic
-
sudo apt-get install libgflags-dev libopencv-dev libboost-all-dev python3.6-dev
python -m venv venv
source venv/bin/activate
# pip install git+https://github.com/ohadmen/pyircp
python3 setup.py install