I am currenty working on a project that involves multiple cameras fixed on the ceiling. Each time I take a picture, I check whether there is a "cart" right under the camera.

I would like to use a Convolutional Neural Network (binary) in order to determine if the image contains a cart or not.

Now, what I am wondering is, since all the carts look exactly the same from the camera's top-down view and that all the images also look pretty much exactly the same (some lighting and slight angle differences but that's it.)

Will this poor image diversity give me any problems ? (such as overfitting) ? If so, what could I do to prevent it ? And.. Do I even want to prevent overfitting in this case ? Since even the real data is going to be pretty much the same.


1 Answer 1


If all of the carts images are similar (and different from the images without carts), the classification problem is easy. Therefore, it is not a problem, but an advantage.

That is under the assumption that the images' distribution in inference is the same distribution as in your training.

  • $\begingroup$ Ok, and would it be correct to assume that since they are all similar, the model could overfit quite easily ? I could I prevent this (if necessary at all ) ? $\endgroup$
    – D.Gaulin
    Commented Jan 12, 2022 at 22:37
  • $\begingroup$ I think it will easilly fit, but it would not be overfit. $\endgroup$ Commented Jan 13, 2022 at 18:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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