SA-MixNet

First, pip install opencv-python == 4.6.0.66 and replace the \anaconda3\envs\pytorch\Lib\site-packages\skimage\segmentation\slic_superpixels.py by the slic_superpixels.py in this repository, to get Road Seed Guided SLIC.
Run road_label_propagation.py to generate proposal masks.
Run train_mix_D.py for training and run test.py for testing.

Name		Name	Last commit message	Last commit date
Latest commit History 9 Commits
SAMix		SAMix
data/DeepGlobe_tiny/train		data/DeepGlobe_tiny/train
data_utils		data_utils
LICENSE		LICENSE
README.md		README.md
RoadPaste_batch.py		RoadPaste_batch.py
cutout.py		cutout.py
graph.py		graph.py
road_label_propagation.py		road_label_propagation.py
skeleton.py		skeleton.py
slic_superpixels.py		slic_superpixels.py

Provide feedback