Skip to content

Latest commit

 

History

History
50 lines (39 loc) · 1.86 KB

installation.md

File metadata and controls

50 lines (39 loc) · 1.86 KB
Raw

Installation Instruction

Start by cloning the repo:

git clone https://github.com/GAP-LAB-CUHK-SZ/SAMPro3D.git
cd SAMPro3D

First of all, you have to make sure that you have all dependencies in place. The simplest way to do so, is to use anaconda.

You can create an anaconda environment called sampro3d as below. For linux, you need to install libopenexr-dev before creating the environment.

sudo apt-get install libopenexr-dev # for linux
conda create -n sampro3d python=3.8 (recommended python version >= 3.8)
conda activate sampro3d

Step 1: install PyTorch (we tested on 1.13.0, but the other recent versions should also work):

pip install torch==1.13.0+cu117 torchvision==0.14.0+cu117 -f https://download.pytorch.org/whl/torch_stable.html

Step 2: build pointops:

cd libs/pointops
# usual
python setup.py install
# docker & multi GPU arch
TORCH_CUDA_ARCH_LIST="ARCH LIST" python setup.py install
# e.g. 7.5: RTX 3000; 8.0: a100 More available in: https://developer.nvidia.com/cuda-gpus
TORCH_CUDA_ARCH_LIST="7.5 8.0" python setup.py install
cd ..

Step 3: install the Segment Anything Model:

pip install git+https://github.com/facebookresearch/segment-anything.git

Then download the default ViT-H SAM model and put it into the folder of SAMPro3D.

(Optional: you may alternatively use the smaller model ViT-L SAM model or ViT-B SAM model by simply changing the args.model_type and args.sam_checkpoint when running our program.)

Step 4: install all the remaining dependencies:

pip install -r requirements.txt