add viser viewer for training, with background image and point cloud viz#135
Conversation
tangkangqi
force-pushed
the
darcyt/viser_train_viewer
branch
2 times, most recently
from
5b3b933 to
9888c86
Compare
tangkangqi
force-pushed
the
darcyt/viser_train_viewer
branch
from
9888c86 to
62910bc
Compare
wilsonCernWq
left a comment
wilsonCernWq
left a comment
There was a problem hiding this comment.
Hi, thanks for the nice contribution, I tested the code and I think I can see point cloud renderings correctly, but I am unable to see, for example, rendered colors. I am wondering if you could take a look and see if something merged recently caused the conflict?
Thank. you!
|
Just noticed this, I will check and update tomorrow. |
tangkangqi
force-pushed
the
darcyt/viser_train_viewer
branch
from
733b413 to
62910bc
Compare
|
Hi @wilsonCernWq , sorry my PC reinstalled thus late... 
|
|
And I add:
|
Updated camera-to-world conversion logic by using the view matrix directly, questioning the previous inversion method.
wilsonCernWq
left a comment
wilsonCernWq
left a comment
There was a problem hiding this comment.
I added one small bug fix, I think the camera transformation is wrong? That is the reason why I saw black images. It seems the issue is obvious on synthestic data, but less obvious on colmap scenes.
…viz (nv-tlabs#135) * add viser viewer for training conifg * add viser gui utils (for training viz), with point cloud and background image option * add resolution adjust with browser size or sidebar config * clear background when render checkbox unchecked * add render of density distance hits normals * Fix camera-to-world conversion logic Updated camera-to-world conversion logic by using the view matrix directly, questioning the previous inversion method. --------- Co-authored-by: tangkangqi <tangkangqi@yeah.net> Co-authored-by: Qi Wu <wilson.over.cloud@gmail.com>
add viser viewer for training, with background image and point cloud viz (nv-tlabs#135) (nv-tlabs#187) See merge request nrs/3dgrut-internal!275
3 participants
Add config param "with_viser_gui"
example command:
python train.py --config-name apps/colmap_3dgut.yaml path=data/mipnerf360/bicycle out_dir=runs experiment_name=bicycle_3dgut dataset.downsample_factor=2 export_usdz.enabled=true with_viser_gui=true
Point cloud update while training:
Background image update while training:
