Detectron2 anyone?

[WaterKnight (David Lacalle Castillo)]

Instance segmentation

Hey guys, The next code is for object detection: pascal = DataBlock(blocks=(ImageBlock, BBoxBlock, BBoxLblBlock), splitter=RandomSplitter(), get_items=get_train_imgs, getters=getters, item_tfms=item_tfms, batch_tfms=batch_tfms, n_inp=1) What do I need to change for instance segmentation? In instance segmentation each box has a mask In adittion, torchvision.models.detection.maskrccn_resnet5…

It is working for variable images sizes. All the work is done in a subclass of Learner.