Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
张佳预 2008bc28d9 | 1 year ago | |
---|---|---|
components | 1 year ago | |
configs | 1 year ago | |
visuals | 1 year ago | |
.gitignore | 1 year ago | |
README.md | 1 year ago | |
dataset.py | 1 year ago | |
requirements.txt | 1 year ago | |
test.py | 1 year ago | |
train.py | 1 year ago |
./config
folder, adjust the hyperparameters as per your needbasepath
)pip install -r requirements.txt
Left (RGB) | Fake Left (NIR) | Right (NIR) | Fake Right (RGB) | Disparity |
---|---|---|---|---|
Method | Common | Light | Glass | Glossy | Veg | Skin | Clothing | Bag | Mean |
---|---|---|---|---|---|---|---|---|---|
From paper (STN + SMN) | 1.13 | 1.55 | 1.05 | 1.52 | 0.89 | 1.23 | 1.14 | 0.98 | 1.18 |
This Implementation | 0.64 | 1.51 | 1.12 | 1.93 | 0.70 | 1.12 | 1.14 | 1.12 | 1.17 |
Comparing STN + SMN since the spectral transalation in this implementation is not F-cyclegan but the original cyclegan.
./weights
according to the default config file)[epoch]_net_G_A.pth
and latest_net_G_A.pth
which signifies the latest checkpoint, you can specify the epoch you want to load weights from in the config file.warmup: True
in config filewarmup: False
for training combined step 1,2,3,4pittsburgh_test.yaml
) and run python test.py --config ./configs/pittsburgh_test.yaml
./summary
foldertensorboard --logdir ./summary
cycleGan跨光谱立体匹配网络
Python other
Dear OpenI User
Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.
For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》