I have been trying to find the test set labels for the retinal vessel segmentation dataset DRIVE. However, all I can find officially is the training labels and all the images, and to evaluate a model, I need to upload my model's predictions and I receive the dice coefficient. However, I need to evaluate my model against other metrics, to have a ground of comparison with other models in the literature. If someone could point me to a link with the test-set ground truth images, I would greatly appreciate it.

Similar questions and discussions