Selective Guided Adversarial Adaptation for Cross-device OCT Images Segmentation
The yellow and the green arrows indicate the source and target domain paths, respectively. The black dashed arrows denote the selective parameter guidance, and the red dashed arrows represent the adversarial learning.
Using the train_SGAA.py
and test.py
to train and test the model on your own dataset,respectively.
The information about the COLA dataset could be seen in the following link: https://imed.nimte.ac.cn/cola.html