5
$\begingroup$

Has anyone end-to-end trained a single net that can find the location of objects in an image (with v11.1)? Here's an example of a region-cnn in action, and this is a good description of the approach.

$\endgroup$
1
$\begingroup$

Not a complete answer, but I think the newly introduced spatial transformer layer can be used to conduct roi pooling, with 0 as the inputs to the skew x and y parameters. I can't think of any other way to do the roi pooling step within Mathematica at the moment, but I haven't fully explored all the new layers yet either.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.