I’m currently looking for a library that can solve or assist solving the following problem.
I have a bunch of 3D point clouds from different angles and would like to combine them to have a single 3d point cloud. Seems like there were some packages like PCL.jl and https://github.com/FugroRoames/PointCloudMatching.jl which seem to be discontinued.
Is there any package that is related to that field that I can have a look at?
I use ImageProjectiveGeometry for a RANSAC implementation to align them in one dimension (detecting the ground and removing it) now I’m mostly interested in rotation on the
z=0 plane as well as translation.
Additionally I’m interested in algorithms in general for that such like
Iterative closest point - Wikipedia.
to maybe be able to create my own package out of it if there doesn’t exist one yet.
Thanks in advance!
Hi @Wikunia , we also have an interest in the topics you mentioned over by Caesar.jl. I recently implemented the first step to easy Point Clouds in Julia, see here ROS → PCL → Caesar._PCL, and the tests I’ve built so far.
I’m sure some bugs will show up, but any community help to give snippets of test data or PRs would be welcome.
This repo contains an example implementation of the iterative closest point (ICP) algorithm. It could / should be factored out into a proper repo and there is still some room for improvement (e.g. usage of StaticArrays).
Ah, thanks for pointing that out! I did some of the StaticArray upgrades by combining that with the known PCL types added to Caesar.jl. I expect to do a few more upgrades so that the transforms are well integrated existing transformation libraries. In the mean time, here is a packaged version of