vicdxxx / clip-iqa Goto Github PK

# Create a conda environment and activate it
conda create -n clipiqa python=3.8 -y
conda activate clipiqa
# Install PyTorch following official instructions, e.g.
conda install pytorch=1.10 torchvision cudatoolkit=11.3 -c pytorch
# Install pre-built MMCV using MIM.
pip3 install openmim
mim install mmcv-full==1.5.0
# Install CLIP-IQA from the source code.
git clone [email protected]:IceClear/CLIP-IQA.git
cd CLIP-IQA
pip install -r requirements.txt
pip install -e .

python demo/clipiqa_koniq_demo.py

python demo/clipiqa_liveiwt_demo.py

# Support dist training as MMEditing
python tools/train.py configs/clipiqa/clipiqa_coop_koniq.py

python demo/clipiqa_koniq_demo.py --config configs/clipiqa/clipiqa_coop_koniq.py --checkpoint ./iter_80000.pth

@inproceedings{wang2022exploring,
    author = {Wang, Jianyi and Chan, Kelvin CK and Loy, Chen Change},
    title = {Exploring CLIP for Assessing the Look and Feel of Images},
    booktitle = {AAAI},
    year = {2023}
}