I was curious if it is possible to create a U-net type architecture that takes in a 3d image and outputs a 2d image? Or, alternatively, would some other architecture be better suited for this problem? Has anyone tried stacking 2d conv layers in keras (or any other lib) on top of 3d conv layers? I am about to try this myself, but was just wondering if it makes sense, and if any deep learning experts cared to weigh in.
$\begingroup$
$\endgroup$
Add a comment
|
