Lane detection is an essential task in understanding the driving corridor for self-driving cars. Precise and accurate lane detection enables self-driving cars to remain centred in lane and track the curves of the road. To ensure safety, lane detection systems must be robust and generalisable across challenging weather conditions, lighting conditions and camera viewing/mounting angles. LaneDet aims to address these challenges through a neural network approach to lane detection as well as lane classification.