A Dataset for Semantic Scene Understanding using LiDAR Sequences


is based on the KITTI Vision Benchmark and we provide semantic annotation for all sequences of the Odometry Benchmark. Overall, we provide an unprecedented number of scans covering the full 360 degree field-of-view of the employed automotive LiDAR.


We labeled each scan resulting in a sequence of labeled point clouds, which were recorded at a rate of 10 Hz. This enables the usage of temporal information for semantic scene understanding and aggregation of information over multiple scans.


We annotated moving and non-moving traffic participants with distinct classes, including cars, trucks, motorcycles, pedestrians, and bicyclists. This enables to reason about dynamic objects in the scene.


Dez 06, 2019 Added semantic scene completion task, code and competition (link).
Aug 19, 2019 Competition for semantic segmentation (link) online and release of the point cloud labeling tool (link).
Jul 15, 2019 Release of dataset including instance annotation for all traffic participants (static and moving).


See also our paper for more information and baseline results:

If you use our dataset or the tools, it would be nice if you cite our paper (PDF):

  author = {J. Behley and M. Garbade and A. Milioto and J. Quenzel and S. Behnke and C. Stachniss and J. Gall},
  title = {{SemanticKITTI: A Dataset for Semantic Scene Understanding of LiDAR Sequences}},
  booktitle = {Proc. of the IEEE/CVF International Conf.~on Computer Vision (ICCV)},
  year = {2019}

But also cite the original KITTI Vision Benchmark:

  author = {A. Geiger and P. Lenz and R. Urtasun},
  title = {{Are we ready for Autonomous Driving? The KITTI Vision Benchmark Suite}},
  booktitle = {Proc.~of the IEEE Conf.~on Computer Vision and Pattern Recognition (CVPR)},
  pages = {3354--3361},
  year = {2012}