search

chensjtu / GaussianObject

GaussianObject: High-Quality 3D Object Reconstruction from Four Views with Gaussian Splatting (SIGGRAPH Asia 2024, TOG)
914 stars 56 forks source link

unrecognized argument #62

Open 7rwang opened 2 weeks ago

7rwang commented 2 weeks ago

Hi everyone,

I'm just a fresher to 3DGS, so I'm not sure why I encounterede this issue train_gs.py: error: unrecognized arguments: -m output/gs_init/kitchen -r 4 --init_pcd_name visual_hull_4 --white_background.

The training order of ''Corase 3DGS'' should be:

1.step sh python train_gs.py -s data/mip360/kitchen \ -m output/gs_init/kitchen \ -r 4 --sparse_view_num 4 --sh_degree 2 \ --init_pcd_name visual_hull_4 \ --white_background --random_background

2.step python render.py \ -m output/gs_init/kitchen \ --sparse_view_num 4 --sh_degree 2 \ --init_pcd_name visual_hull_4 \ --white_background --skip_all --skip_train, right?

BTW do we need mkdir ''output/gs_init/kitchen'' manually?

UranusITS commented 2 weeks ago

The \ symbol in the README is used for line breaks to improve readability. If you want to run the code in a single line, please just execute:

python train_gs.py -s data/mip360/kitchen -m output/gs_init/kitchen -r 4 --sparse_view_num 4 --sh_degree 2 --init_pcd_name visual_hull_4 --white_background --random_background