├── GLNet_Code.zip
└── README.md


/GLNet_Code.zip:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/rmcong/GLNet_TCYB2022/HEAD/GLNet_Code.zip


--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
 1 | # GLNet_TCYB2022
 2 | Runmin Cong, Ning Yang, Chongyi Li, Huazhu Fu, Yao Zhao, Qingming Huang, and Sam Kwong, Global-and-local collaborative learning for co-Salient object detection, IEEE Transactions on Cybernetics, 2022.
 3 | 
 4 | Project: https://rmcong.github.io/proj_GLNet.html
 5 | 
 6 | # Results of GLNet:
 7 | * Results:
 8 |   - We provide the resutls of our GLNet on Cosal2015, iCoseg, and MSRC. 
 9 | ```
10 | Baidu Cloud: https://pan.baidu.com/s/1sXBc4H3fKK8Y8ceaU4AjSQ  Password: 0224
11 | ```
12 | 
13 | 
14 | # Pytorch Code of GLNet:
15 | * Pytorch implementation of GLNet
16 | * Pretrained model:
17 |   - We provide our testing code. If you test our model, please download the pretrained model, unzip it, and put the checkpoint `model_GLNet.pth` to `Checkpoints/trained/` folder 
18 |   and put the pretrained backbone `backbone_v.pth` to `Checkpoints/warehouse/` folder.
19 |   - Pretrained model download:
20 | ```
21 | Baidu Cloud: https://pan.baidu.com/s/1sXBc4H3fKK8Y8ceaU4AjSQ  Password: 0224
22 | ```
23 | 
24 | ## Requirements
25 | 
26 | * Python 3.7
27 | * Pytorch 1.5.1
28 | * torchvision
29 | 
30 | ## Data Preprocessing
31 | * We resize the images of original test datasets. Please download the resized data, and put the data to `Data/` folder.
32 | * Resized test datasets:
33 | ```
34 | Baidu Cloud: https://pan.baidu.com/s/1sXBc4H3fKK8Y8ceaU4AjSQ  Password: 0224
35 | ```
36 | 
37 | ## Test
38 | ```
39 | python test.py
40 | ```
41 | 
42 | * You can find the results in the `'Outputs/'` folder.
43 | 
44 | # If you use our GLNet, please cite our paper:
45 | 
46 |     @article{GLNet,
47 |         title={Global-and-local collaborative learning for co-Salient object detection},
48 |         author={Cong, Runmin and Yang, Ning and Li, Chongyi and Fu, Huazhu and Zhao, Yao and Huang, Qingming and Kwong, Sam},
49 |         journal={IEEE Trans. Cybern.},
50 |         year={early access, doi: 10.1109/TCYB.2022.3169431},
51 |         publisher={IEEE}
52 |     }
53 | 
54 | # Contact Us:
55 | If you have any questions, please contact Runmin Cong (rmcong@bjtu.edu.cn) or Ning Yang (ningyang@bjtu.edu.cn).
56 | 


--------------------------------------------------------------------------------