It seems that to train this, you need to input precise masks.
I'm wondering if one day it would be possible to train a network without masks (just a classifier), and it will figure out the masks by itself.
TensorFlow Mask R-CNN code for pixelwise object detection and segmentation
41–43 of 43 posts
Re: TensorFlow Mask R-CNN code for pixelwise object detection and segmentation
#42Earlier quoted context omitted.
Yes, and that's possible now with many different CNNs. The limiting factor is the training/validation/test data in your subject. For example you couldn't implement Mask R-CNN with the COCO dataset as implemented here and get inference on your radiology problem set.
Would transfer learning help with that?
Yea I mean transfer learning would bring over the first n convolutions so it would be faster than from scratch, but you still need the radiology data to get the last few steps.
Re: TensorFlow Mask R-CNN code for pixelwise object detection and segmentation
#43SUPER EXCITING!