About using edge boxes

The question is about using Edgeboxes.
In the sample of Edgeboxes, a model is required as the input to create at the very beginning, see line below.
cv::Ptrcv::ximgproc::StructuredEdgeDetection pDollar = cv::ximgproc::createStructuredEdgeDetection( model );
What I don’t know is simply the ‘model’. What the model is and how can I get it?
It seems to be something trainned in advance according to opencv_contrib/structured_edge_detection.cpp at 4.x · opencv/opencv_contrib · GitHub

(dont unzip it, it is meant to be used as-is)