What Do Deep Saliency Models Learn about Visual Attention?

Open in new window