Please follow list to put the BDD100K dataset (train, val, test) in the desired folder. The labels generated by our method are provided here (train and val). We'll call the directory that you cloned ENet-BDD100K-Torch as $ENet_BDD100K_ROOT. Note that this model only uses ENet as backbone, and if you use ENet-SAD, the performance will be better.
- Run test script
By now, you should be able to reproduce the result (Accuracy: 0.3656, mIoU: 16.02).
cd $ENet_BDD100K_ROOT sh ./experiments/test_ENet.sh
- Training ENet model
The training process should start and trained models would be saved in
cd $ENet_BDD100K_ROOT sh ./experiments/train_ENet.sh
experiments/models/ENet-new/by default.
Then you can test the trained model following the Testing steps above. If your model position or name is changed, remember to set them to yours accordingly.