We have started work on collecting bounding box annotations for our dataset, while our bounding-box labeled dataset maybe small. I think we should start thinking about implementing Object segmentation networks for the same. We could consider Fast(er)-RCNN, YOLOv3 and SSDs and check which performs the best.
I think this issue can be used for discussion about the same.
We have started work on collecting bounding box annotations for our dataset, while our bounding-box labeled dataset maybe small. I think we should start thinking about implementing Object segmentation networks for the same. We could consider Fast(er)-RCNN, YOLOv3 and SSDs and check which performs the best. I think this issue can be used for discussion about the same.