Learning-Based Visual Saliency Model for Detecting Diabetic Macular Edema in Retinal Image

Figure 3

Some saliency maps produced by 9 different models from the EDMERI database along with predictions of several models using ROC. Each example shown by one row. From left to right: original image, ground truth, Ours, AIM, AWS, GBVS, ITTI, STB, Judd, SUN, and Torralba. It is obvious that Ours is more similar to the ground truth than other saliency maps.