In this project, the pixels of a road are labeled using a Fully Convolutional Network (FCN).
Make sure you have the following is installed:
Download the Kitti Road dataset from here. Extract the dataset in the data
folder. This will create the folder data_road
with all the training a test images.
Run the following command to run the project:
python main.py
- Adam optimiser
- Learning rate : 1e-4
- Dropout : 50%
- L2 Reg: 1e-3
- Epochs : 30
- Batch size : 16
- FCN-8 network architecture
Here are a couple of great results I was able to achieve: