I fine-tuned the panoptic segmentation model with my custom dataset, where I have two classes (background and object).
I wanted to ask if it is possible to force the number of object masks to be detected on an image? For example, if I know that my image contains 3 objects, I would like to get 1 mask for the background and 3 masks for the objects.
Hello,
I fine-tuned the panoptic segmentation model with my custom dataset, where I have two classes (background and object).
I wanted to ask if it is possible to force the number of object masks to be detected on an image? For example, if I know that my image contains 3 objects, I would like to get 1 mask for the background and 3 masks for the objects.
Thanks!