Is Elon Musk wrong about LiDAR? We compared camera and LiDAR…

Is Elon Musk wrong about LiDAR? We compared camera and LiDAR training data to put the two self driving methods to the test.

Takeaways

  • 2D annotations often have deeper inaccuracies hiding beneath the surface.
  • 2D annotations probably still have a place either as part of a larger sensor system or for simpler tasks such as classifying objects for basic lane-keeping/collision avoidance or highway driving.
  • It’s always better to have multiple sensor modalities available. 
  • The findings support the idea of a virtuous cycle in ML development: more robust sensors lead to much higher accuracy training data which means the perception model will perform better, which in turn may eventually lower reliance on any one sensor type. But the inverse effect exists too: even if a safe self-driving system is theoretically possible without LiDAR, getting great training data will be much tougher with cameras alone. 

https://autonomous-driving-news.tumblr.com/post/187136655449

Leave a Reply