Jetson Inference Segnet Model for Lane detection

Hello,

I am fairly new into the machine learning world and I would like to ask if there is a pretrained network for lane detection,e.g. Lanenet, similar to the other segmentation networks in the Jetson inference package that can be run by the ROS Nodes from NVIDIA from the ROS deep learning package.
If not, what would be the recommended way to train such a network myself?

Thank you

Hi Simon, if you check out the Cityscapes models, those include a class for the roadway. However I don’t believe they explicitly detect different lanes in the same road. You would need to find/create such a dataset and re-train the network on it.

A quick google search turned up this for lane segmentation dataset: https://arxiv.org/abs/1807.01347
Also this is the PyTorch code I use for segmentation training: https://github.com/dusty-nv/pytorch-segmentation