Is there a neural network that has architecture optimizations for segmenting only one class (object and background)? I have tried U-net but it is not providing good enough results.

I am wondering if this can be due to the fact that my dataset has different image resolutions/aspect ratios.

  • $\begingroup$ When you say "only one class (object and background)", in fact you are saying two classes: the object class and the background class. $\endgroup$ Dec 14 '19 at 11:48
  • $\begingroup$ To allow us help you, you should describe a few more the context: kind of images, ... . Also the expected result. $\endgroup$ Dec 14 '19 at 12:02
  • $\begingroup$ @pasabaporaqui Thanks for reply. I'm trying to train a car segmentation. I have custom dataset I've collected. I expect it to have precise boundary around the car on an image. $\endgroup$
    – Dusan J.
    Dec 17 '19 at 9:41

Your Answer

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

Browse other questions tagged or ask your own question.