2

I am trying to build an image classification model with Keras that should be able to classify cups as "good condition" or "defect". This was easy to do with a single image as input. However, I now want to try feeding 6 images, one from every angle(top, bottom, side etc), as input. What would be the best approach for this ? My initial idea was a np array of shape (6, width, height, 3), but this deemed unsuitable.

can_k
  • 31
  • 3
  • why not concate all the images one by one to make a big image and train a CNN model on such images? Or you can make a model with multiple inputs, which feeds in to seperate CNNs and then combine the results from each model – Sreeram TP Sep 16 '21 at 12:58
  • 2
    You have several possibilities to test and check which one gives you more accuracy. You can merge these 6 with 3 channels like (w,h,6*3=18), or (w,h,6,3) as a 3D image and using Conv3D and MaxPool3D, or even feed them as separate samples (w,h,3). – Kaveh Sep 16 '21 at 12:59
  • I’m voting to close this question because it is not about programming as defined in the [help] but about ML theory and/or methodology - please see the intro and NOTE in the `machine-learning` [tag info](https://stackoverflow.com/tags/machine-learning/info). – desertnaut Sep 16 '21 at 17:09

0 Answers0