Is this an end-to-end model?

[WilburZjh]

I have one question: it points out in the paper that the "We collect all the detections produced by DeepDet to train the DeepVer model." May I know how you get the final score of an image patch? The formulation is : "S = ω × SDeepDet + (1 − ω) × SDeepVer". Is this for entire model or for each patch in the training data?

[ChaoLi977]

In the test stage, the final score of an image patch is the formulation as above: S = ω × SDeepDet + (1 − ω) × SDeepVer. While in the training stage, the DeepDet model and DeepVer model are trained independently. The balance weight ω is adjusted on the training set to the best performance.