Open manyaafonso opened 4 years ago
It is assumed that for my data (single class), at the most one object can and should be detected per image.
Multiple objects of the same class are detected in some images.
infer_simple.py
Is there any setting or parameters to force MaskRCNN to learn that there can be at most one object of the single class of interest per image?
Expected results
It is assumed that for my data (single class), at the most one object can and should be detected per image.
Actual results
Multiple objects of the same class are detected in some images.
Detailed steps to reproduce
infer_simple.py
System information
Is there any setting or parameters to force MaskRCNN to learn that there can be at most one object of the single class of interest per image?