We convert the pixel-level saliency maps into the UI element-level saliency maps.
For this, we compute the integral of the pixel-level saliency density over the area
covered by an element. This is followed by normalization over all elements to ensure
that the vector sums to 1. Given an UI with k elements, we represent the element
saliency map E = (E1, ..., Ek)', as vector of probabilities where Ej is the probability
of element j being fixated. In case one UI element overlaps another element, we
assign the saliency of the pixels in overlapping regions to the element on top.
Sample pixel-level and element-level saliency maps are presented in Figure 7.
関連論文 Saliency Prediction for Mobile Interfaces
3.4.2 Element-level Saliency
ToDo