Allow non-square input shape for keras models about coral-deeplab HOT 7 CLOSED

Hi,

I could push update to allow non square input size. There are no pretrained weights for it, so you'll have to train from scratch.

But in your particular case - I'm not sure if this kind of resolution will work with Edge TPU. I gave it a quick try and compiler failed to create a model.

from coral-deeplab.

Sure, I will do experiments
Maybe 800x800 is good solution Than I will divide picture into 2 parts

from coral-deeplab.

I had closer look at this and it seems like edgetpu_compiler is not particularly happy about non square inputs for this model (regardless of shape). Since I want to be sure that people using pretrained and finetuned keras model have no problem compiling it to the device, check for square inputs will not be removed.

from coral-deeplab.

Got it. It is very interesting why...

from coral-deeplab.

Dear @xadrianzetx I noticed before that deeplab recognize some objects better when input is larger. Could you advice how to change from 513x513 to 800x800?

from coral-deeplab.

You can create new instance of keras model with input shape you want like so:

import coral_deeplab as cdl

model = cdl.applications.CoralDeepLabV3(input_shape=(800, 800, 3))

As I said earlier, there are no pretrained weights for this shape, so you have to train this model yourself. I think you can use scripts you posted in other issue to do so.

from coral-deeplab.

FYI: 800x800 trainubg with Nvidia 1080Ti failed, I have switched to try train 512x512

from coral-deeplab.