Closed mdv3101 closed 2 years ago
I want to train a Cascade RCNN with negative images (images which contain only background). I noticed that the model is filtering out the negative dataset at the time of loading annotations into the memory. Is there any way to solve this?
I want to train a Cascade RCNN with negative images (images which contain only background). I noticed that the model is filtering out the negative dataset at the time of loading annotations into the memory. Is there any way to solve this?