Longyong Wu

There are some features like `execution` that have been used. https://github.com/hku-mars/M-detector/blob/0645dabe307a99f3b59a5a3d3007acfdcad1a39f/include/m-detector/DynObjFilter.h#L26 https://github.com/hku-mars/M-detector/blob/0645dabe307a99f3b59a5a3d3007acfdcad1a39f/include/m-detector/DynObjFilter.h#L324-L327 I have tried running it with commenting the usage of `execution` as well as the fast-lio. It seems...

> Thank you so much! By the way, I am testing this code with 16 line LiDAR, and I am confused with the parameters. Can you give me some advices?...

If you need low poly, you can refer to [polygonal surface reconstruction](https://doc.cgal.org/latest/Polygonal_surface_reconstruction/index.html) and [kinetic surface reconstruction](https://cgal.geometryfactory.com/CGAL/doc/master/Kinetic_surface_reconstruction/index.html). If you just need basic mesh, [poisson reconstruction](https://www.open3d.org/docs/latest/tutorial/Advanced/surface_reconstruction.html) like methods might be what you...

The minimal test case is ```python import pickle import numpy as np import open3d as o3d pcd = o3d.io.read_point_cloud(o3d.data.KnotMesh().path) pcd.estimate_normals() pcd.estimate_covariances() print(pcd.has_covariances(), pcd.has_normals()) data = pickle.dumps(pcd) pcd_loaded = pickle.loads(data) print(np.all(np.asarray(pcd.points)...

It seems that you used the old version of `o3d`, you could downgrade the `numpy` version to `1.26.4`

This link https://github.com/hku-mars/M-detector/tree/FAST_LIO