Compute Platform


Machine Learning



  • Deep Virtual Stereo Odometry
  • PythonRobotics
    A collection of Python robotics algorithms with minimum dependencies for autonomous vehicle navigation

Path Planning



  • Comparision of Semantic Segmentation Datasets
  • Kitti Dataset
    First destination for autonomous driving researchers – you’ll find annotated for anything from scene flow to lidar until 3D object localization. The benchmarks on Kitti are a battleground for researchers in pursuit of the sleekest algorithms – this is were you should look when you need reference implementations.
  • Cityscapes
    Daimler’s public dataset: 5k precisely and 20k weakly annotated images. Have a look at the contributions in the benchmark area.
  • Apollo Open Data Platform
    New and exciting: 200k annotated images, several different annotations, definitely worth a visit.
  • Mapillary vistas dataset
    20k images from all over the world, taken by a variety of different cameras. Great for training focusing on robustness against camera type and deployment location.
  • BDD100k
    Probably the largest publicly available self-driving dataset. Includes object detection, Lane detection, drivable area, and semantic instance segmentation sub-datasets.
  • Google Open Images Dataset V4
    _millions_ of images with 2d boundind box and image tag annotations.
  • SYNTHIA Dataset
    Tons of synthetic data from a virtual environment. Interesting for transfer learning.
  • German traffic sign benchmark
    Lives up to its name; the de facto standard for traffic sign recognition.
  • Tsinghua-Tencent 100K
    100k labeled chinese traffic sign images.
  • LabelMe
    Label Me is not only a collaborative labeling tool, but also contains a ton of annotated data in the public dataset section.




  • MIT 6.S094: Introduction to Deep Learning and Self-Driving Cars
  • MIT Self-Driving Cars: Sacha Arnoud, Director of Engineering, Waymo
  • Chris Urmson: How a driverless car sees the road
  • MIT Self-Driving Cars: Sterling Anderson, Co-Founder, Aurora

By continuing to use the site, you agree to the use of cookies. more information

The cookie settings on this website are set to "allow cookies" to give you the best browsing experience possible. If you continue to use this website without changing your cookie settings or you click "Accept" below then you are consenting to this.