Pretrained model Coco with Resnet50 layers
evinpinar opened this issue · 3 comments
evinpinar commented
Currently, there are two pretrained model weights:
- Resnet50 Imagenet weights for feature extraction (backbone)
- MaskRcnn Coco weights for mask-rcnn with resnet101 architecture
Are there any coco trained maskrcnn weights for resnet50 architecture? Or is it okay if I use the resnet101model weights on resnet50 architecture?
evinpinar commented
So far, according to my experiments, it does not work well.
bpmsilva commented
@evinpinar, you could use this tutorial, which uses a native implementation of Mask RCNN in Pytorch
evinpinar commented
Thanks, but I need the weights because i'm modifying this implementation a bit.
I guess i will instead move to fb official implementation.