WHAM Model
์ฃผ์
WSL์์ conda ๋ชจ๋ธ ๋๋ฆฌ๊ธฐ
๋ชฉ์ฐจ | ์ค๋ช | ๋งํฌ |
---|---|---|
ํ๊ฒฝ | Window 11, WSL 2.0 | ๋ฐ๋ก๊ฐ๊ธฐ |
์ฝ๋ | WHAM Installation ์ฝ๋ | ๋ฐ๋ก๊ฐ๊ธฐ |
๊ตฌํ ์ฝ๋ | WHAM Installation ์ฝ๋ | ๋ฐ๋ก๊ฐ๊ธฐ |
ํ๊ฒฝ
- Window 11
- WSL 2.0
์ฝ๋
์๋ณธ Installation ์ฝ๋๋ ์๋์ ๊ฐ๋ค.
github
๋งํฌ
๊ตฌํ ์ฝ๋
# Clone the repo
git clone https://github.com/yohanshin/WHAM.git --recursive
cd WHAM/
# Create Conda environment
conda create -n wham python=3.9
conda activate wham
# Install PyTorch libraries
conda install pytorch==1.11.0 torchvision==0.12.0 torchaudio==0.11.0 cudatoolkit=11.3 -c pytorch
# Install PyTorch3D (optional) for visualization
conda install -c fvcore -c iopath -c conda-forge fvcore iopath
pip install pytorch3d -f https://dl.fbaipublicfiles.com/pytorch3d/packaging/wheels/py39_cu113_pyt1110/download.html
# Install WHAM dependencies
pip install -r requirements.txt
# Install ViTPose
pip install -v -e third-party/ViTPose
# Install DPVO
cd third-party/DPVO
wget https://gitlab.com/libeigen/eigen/-/archive/3.4.0/eigen-3.4.0.zip
unzip eigen-3.4.0.zip -d thirdparty && rm -rf eigen-3.4.0.zip
conda install pytorch-scatter=2.0.9 -c rusty1s
conda install cudatoolkit-dev=11.3.1 -c conda-forge
# ๊ทธ๋ฅ ๋ถ์ฌ๋ฃ๊ธฐ ํ์๋ฉด ๋ฉ๋๋ค.
# ONLY IF your GCC version is larger than 10
conda install -c conda-forge gxx=9.5
pip install .
# ์๊ธด ์ค๋ฅ : undefined symbol: iJIT_NotifyEvent
# ํด๊ฒฐ ์ฝ๋ (์์ธ : mkl ๋ฒ์ ๋ฌธ์ )
conda install mkl==2024.0
pip install .
# Download demo data
# ์ฌ๊ธฐ ๋ค์ด๋ก๋ ๋ฐ๊ธฐ ์ ์ ๋ฐ์ SMPL, SMPLify์ ํ์๊ฐ์
ํด์ผํ๋ค.
# ์ค๊ฐ์ ์์ด๋ ๋น๋ฒ ์น๋ ๊ฒ์ด ์์ (๋ฑ๋ก ์ํ๊ณ ํ๋ฉด ๋ชจ๋ธ ๋ค์ด๋ก๋ ์๋จ..)
bash fetch_demo_data.sh
๋๋ฆฌ๋ค๊ฐ ๊ทธ๋ง ๋์๋ค.
- ์๋ํ๋ฉด, ๋ชจ๋ธ์ด ๋๋ฌด ๋ฌด๊ฑฐ์์, 8GB GPU์์๋ ์ถ๋ก ์กฐ์ฐจ ์๋์๋ค.
- Github Issue ๋ค์ด๊ฐ๋ณด๋ฉด ์ ์ ์๋ค.