LLAMAS is an Automatically annotated lane markers using Lidar maps for labelled lane markers, containing over 100 000 annotated images, covering over one kilometre of road footage. Lane markers are tricky to annotate because of their median width of only 12 cm. At farther distances, the number of pixels gets very sparse and the markers start to blend with the asphalt in the camera image. While pixel-level segmentation can be very useful for localization, some automated driving systems benefit from higher level representations such as splines, clothoids, or polynomials. This section of the dataset allows for evaluating existing and novel techniques.
For more information please read Unsupervised Labeled Lane Markers Using Maps.
Detailed information about the annotations can be consulted here.
The dataset can be downloaded here. You must create an account and log in in order to download the dataset.
Simple baseline models have been provided. You can check those in the benchmarks section. Some improved models are present in the benchmarks, but not all of them have a published code.
You can consult the binary lane marker segmentation, the multiclass segmentation or the lane approximation benchmarks.
Dataset licensed under a Non-Commercial license.