py4dgeo
py4dgeo copied to clipboard
3dgeo-heidelberg
py4dgeo - A Python library for change analysis in 4D point clouds
In addition to the plot() method, it would be nice to have an animate() method for objects of 4D-OBC extraction. As we know, it is helpful to understand the extracted...
The first step of the correspondence-driven plane-based M3C2. As the first iteration, this will be used afterward for testing comparison ( e.g. correctitude and performance improvement )
In the original method seeds that are not finished are not used for region growing. We should make it possible for online settings, but they should be skipped as a...
Using https://github.com/fatiando/pooch to download the test data will allow us to cache the data properly.
This is a follow-up to #161: For very large analysis that are built incrementally, all algorithms should be able to build on previous results and only extend it for new...
We are currently not interpolating between elements to determine the 95th percentile. This is not super-important, but should be fixed at some point.
With the help of this [git issue](https://github.com/ssciwr/py4dgeo/issues/84) I'm able to run M3C2 algorithm with (I think) the same params used in CloudCompare. However, the M3C2 algorithm takes roughly 100x times...
This allows an actual pure Python implementation that a user would expect from `PythonFallbackM3C2`.
This is a rough idea that could be worth exploring to increase the cache locality of our computations: `KDTree` internally stores a vector of point indices. This index vector could...
Performance benchmarking shows that as is, the KDTree construction is the bottleneck of the entire algorithm. A naive approach to parallelization could be to distribute the subtree construction to threads....
Metadata
Owner
Metadata
py4dgeo - A Python library for change analysis in 4D point clouds