Hi,
Thanks for your excellent work, here I wanna know about the prior map, actually, maybe you are the first to propose the learned priors, different from the traditional methods which used the hand-crafted prior maps.
However, rather than knowing how to use the prior maps, I'd like to know more about how you get the prior maps. What does the learn mean in learned priors? Is there another network to learn to generate the prior maps? Or is there any ground truth to train the network to produce the prior maps?
Hi, Thanks for your excellent work, here I wanna know about the prior map, actually, maybe you are the first to propose the learned priors, different from the traditional methods which used the hand-crafted prior maps.
However, rather than knowing how to use the prior maps, I'd like to know more about how you get the prior maps. What does the learn mean in learned priors? Is there another network to learn to generate the prior maps? Or is there any ground truth to train the network to produce the prior maps?
Looking forward to your reply!