Correct architecture for lots of small objects?

[fschi]

Hello, I have a question regarding the architecture. I have images like this example, and want to do image segmentation for the white spots.

Do I understand it correctly, that in your architecture there are no limitations about the number of objects that can be detected? Are there limitations about the size of the objects? If not, I think for a beginner it would be a good start to use your architecture...

Thanks for a short answer!

[petrosgk]

Hello, You are correct that the architecture can be used for any segmentation task. In fact, the U-Net architecture was first used for biomedical image segmentation.

[fschi]

Ok thanks! but is there a limit about the number of different areas that can be detected? In the examples I’ve found most of the times there are less than 10 areas. So I’ve wondered if there is a limit inside the architecture..

[petrosgk]

The limiting factor is that the model must be expressive enough to solve a task with an accuracy you deem acceptable. If the architecture doesn't perform as well as you need in your segmentation task then you'll have to improve upon it or use a different architecture.