This repository is to take CVPR 2018 WAD Video Segmentation Challenge with SegNet.
Follow the instruction to download dataset.
$ kaggle competitions download -c cvpr-2018-autonomous-driving -p /mnt/data
Download VGG16 into models folder.
Extract training images:
$ python pre-process.py
$ python train.py
If you want to visualize during training, run in your terminal:
$ tensorboard --logdir path_to_current_dir/logs
Download pre-trained SegNet Model to "models" folder then run:
$ python demo.py
Then check results in images folder, something like:
Input | GT | Output |
---|---|---|
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
Legend: