Suitable instance counting CNN for training on polygonal masks

Question

I have a medical dataset labeled with polygonal masks (rather than rectangle boxes). It works well for pixel annotation with UNet to generate masks of healthy vs damaged skin. Now I need to do instance counting. Most of the CNNs like YOLOv4 consumes bounding boxes, not polygons.

Which CNN should I use for instance counting given that my dataset consists of labeled polygons?

Answer 1

Which CNN should I use for instance counting given that my dataset consists of labeled polygons?

CNNs for instance segmentation. To start with you can try Mask-RCNN. Here are all state of the art CNNs for instance segmentation. You will also find code for most of them.