4 citations · 10 across the 7 of their papers we have counts for
1 paper · 1 filter
Yutong Sun, Mohit Prabhushankar, Ghassan AlRegib
In this paper, we show that existing recognition and localization deep architectures, that have not been exposed to eye tracking data or any saliency datasets, are capable of predi…