You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Thank you for your great work for hierarchical 3DGS!
After training with wayve dataset, the result shows great as follows:
When I look into the traing_chunk in details, it shows there are some intermidiate point_cloud result in output/trained_chunks/0_0. when I tried to load it in original 3DGS viewer, but it shows some artifacts(e.g. some white lines near the wheel).
The loading command is: SIBR_gaussianViewer_app -m ./output/trained_chunks/0_0. it seems the opacity is not translated well in the original 3DGS viewer.
So My question is: what's the main difference between the original ply format and the new ply format in hierarchical-3d-gaussians? Thanks!
The text was updated successfully, but these errors were encountered:
thanks for trying! Yes, there is a key difference: the new viewer uses correct antialiasing. This is something that will be backported to the original 3DGS viewer soon as well, so you should be able to view them correctly.
thanks for trying! Yes, there is a key difference: the new viewer uses correct antialiasing. This is something that will be backported to the original 3DGS viewer soon as well, so you should be able to view them correctly.
Hth! Bernhard
Thank you, Bernhard! it's related to DGR_FIX_AA settings in the code.
Thank you for your great work for hierarchical 3DGS!
After training with wayve dataset, the result shows great as follows:
When I look into the traing_chunk in details, it shows there are some intermidiate point_cloud result in output/trained_chunks/0_0. when I tried to load it in original 3DGS viewer, but it shows some artifacts(e.g. some white lines near the wheel).
The loading command is: SIBR_gaussianViewer_app -m ./output/trained_chunks/0_0. it seems the opacity is not translated well in the original 3DGS viewer.
So My question is: what's the main difference between the original ply format and the new ply format in hierarchical-3d-gaussians? Thanks!
The text was updated successfully, but these errors were encountered: