Learning to Model Task-Oriented Attention

Figure 4

Percent of gaze points that fall within the target object and the center area.