search

aharley / segaware

Segmentation-Aware Convolutional Networks Using Local Attention Masks
145 stars 31 forks source link

How to train embedding network #5

Open jinde-liu opened 6 years ago

jinde-liu commented 6 years ago

I have found that the code only contains the part for test and no code for training. Is the embedding network trained separately from deeplab network?

aharley commented 6 years ago

Yes: I recommend pre-training the embeddings using VOC 2012, or COCO. You can also train them end-to-end in the full pipeline. I provided a config file for an embedding network.