Open mcmingchang opened 1 month ago
I need to rewrite the algorithm into a GPU version for the project. I have already completed voxelization and normal vector calculation, but I still need the final ICP and GICP. I found that voxelization involves adding a fixed value, which may cause problems if it is smaller than the minimum value of the point cloud.
May I ask if the gicp implementation method of the project is similar to that of open3d? Why do I see some slight differences from the source code perspective?