A review of lane detection methods based on deep learning

Highlights：

• This work is the _rst overall review of recent deep learning-based lane detection methods.

• Detailed description of representive methods from perpective of computer vision and pattern recognition.

• Detailed description of convolution neural networks' architectures and loss functions that used in lanes detector.

• Advantages of deep learning-based methods compared with traditional heuristic recognition-based methods.

• Current challenges of existing deep learning-based methods and some possible directions to solve the problems.