LaneDetection: spatialCNN (SCNN) - Implement/Test lane detection (open source dataset)

[Christiaan1990]

Goal:

Implement, train and test a spatialCNN algorithm and test it with the open source datasets CULane and TuSimple

Assigments:

• Understand how the ML-algorithms works
• Implement the ML-algorithms
• Train and test it with an open source dataset
• Documentation in the TUM-Wiki: https://wiki.tum.de/display/phoenix/Machine+Learning

Notes:

• https://paperswithcode.com/task/lane-detection/codeless
• See approach 2 of https://towardsdatascience.com/tutorial-build-a-lane-detector-679fd8953132
• optional dataset: https://xingangpan.github.io/projects/CULane.html
• optional dataset: tusimple benchmark ground truth TuSimple/tusimple-benchmark#3
• basic information about lane detection: https://wiki.tum.de/display/phoenix/Lane+Detection

[MaslinuPoimal]

Note: this might be a helpful link.

In particular, we can try out the top paper which seems to be speed-oriented as well.

[Hoff97]

A good working implementation of SCNN in pytorch can be found here:
https://github.com/harryhan618/SCNN_Pytorch

[Hoff97]

I tested a model trained on TuSimple on some simple images from the car. It doesnt look like the network can detect any lanes, probably because the training pictures look too different.