下载源码:git clone --recurse-submodules https://github.com/graphdeco-inria/gaussian-splatting.git
利用conda创建虚拟环境:
conda create -n gs python=3.8
切换虚拟环境:
conda activate gs
在gs环境下安装:
pip install torch==2.0.0+cu118 torchvision==0.15.0+cu118 torchaudio==2.0.0+cu118 -f https://download.pytorch.org/whl/torch_stable.html
pip install plyfile
pip install tqdm进
入项目文件夹:
cd gaussian-splatting
pip install submodules/diff-gaussian-rasterization/
pip install submodules/simple-knn/
训练:
python train.py -s data/tandt/train
python train.py -s data/db/playroom