What Do Deep Saliency Models Learn about Visual Attention?

FAtt
Re-assign community