-
Notifications
You must be signed in to change notification settings - Fork 2.6k
Open
Description
Change SiftExtraction.use_gpu
to FeatureExtraction.use_gpu
in convert.py
doesn't work for me, the implementing result is stuck at feature extraction process.
I installed colmap and imagemagick:
#install colmap
!sudo apt update !sudo apt install -y colmap
#install imageMagik
!sudo apt install -y imagemagick
Then put all the images in folder: /content/input
, and change directory to gaussian-splatting %cd /content/gaussian-splatting
.
Finally convert image: !python convert.py -s /content/input
.
When use FeatureExtraction.use_gpu
in convert.py
, the error shows can't recognize FeatureExtraction:
ERROR: Failed to parse options - unrecognised option '--FeatureExtraction.use_gpu'.
ERROR:root:Feature extraction failed with code 256. Exiting.
When use SiftExtraction.use_gpu
in convert.py
, the error shows can't find image path:
[option_manager.cc:811] Check failed: ExistsDir(*image_path)
ERROR: Invalid options provided.
ERROR:root:Feature extraction failed with code 256. Exiting.
Does anyone know how to fix it?
Metadata
Metadata
Assignees
Labels
No labels