'Rethinking Atrous Convolution for Semantic Image Segmentation'

1. Pre-traind Models


deeplabv3-voc2012.pt	- Trained on VOC 2012 training set for 54,000 steps. - Average mIoU on VOC 2012 validation set: 0.6563

Ground truth	Prediction

Ground truth	Prediction

Name		Name	Last commit message	Last commit date
Latest commit History 88 Commits
papers		papers
scripts		scripts
voc2012_val_predictions		voc2012_val_predictions
.gitignore		.gitignore
README.md		README.md
model.py		model.py
predict.py		predict.py
train.py		train.py
utils.py		utils.py
voc2012.py		voc2012.py