poor result
jhy-dabeeo opened this issue · 4 comments
Hi
I'm trying to make building models using your project.
but cannot get a decent result like your examples
can i get any advice or tip?
this is my input pointcloud building
these are RANSAC parameters ( all default )
before detect Planars, i run Estimate normals with deafult setting
this is the RANSAC result. it looks good to me
After saving the ply and vg files, run make_pc_dataset.py script and reconstruct.py
all parameters in config.yaml are default except normalize and append_bottom
cell_complex.visualize() result
this is final result
All structures on the roof of the building have disappeared.
Hi @jhy-dabeeo, three possible causes on top of my head:
- The default
coefficient
value may not fit your data. Try to decrease the value to avoid shrinking surfaces. The effect of this parameter is explained in section 5.2.4 of the paper. - Something possibly cracked with cell complex creation. From the
cell_complex.visualise()
screenshot I can't really see the cells from the slanted roof structure. Can you make sure they're there? You can also save the cells into PLM files withcell_complex.save_plm()
, load them into Mapple, and switch on clipping-plane for better visualisation (Global - Clipping Plane). - Did you use the provided Point2Surf model for SDF prediction? This pre-trained model may not fit your data (though it gets the overall building shape), in which case you need to retrain the model.
Thanks for the follow-up. Now I can see the slanted roof structure in the cell complex. Your point cloud seems very noisy, which could result in
- excessive cells being generated from excessive plane segments,
- the model pre-trained only with Gaussian noise being confused.
In this case, I'd guess possibly the combination of 1 and 2 resulted in the poor reconstruction result. Maybe try to tune the plane detection further (IMO this is actually the main limitation: in the paper we assumed good-quality planes can be detected), or retrain a model with data of similar characteristics?
Closed for now. Feel free to reopen in case of any updates.