Genau genommen nicht weniger als der Mensch. Nur, dass der das offenbar anders verarbeitet. Schon beeindruckend. Bei beiden.
This video shows what Tesla Autopilot’s neural network sees on the road. Tesla says the system relies on per-camera networks to analyze raw images to perform semantic segmentation, object detection and monocular depth estimation. It employes birds-eye-view networks to take a video from all cameras to output the road layout, static infrastructure and 3D objects directly in the top-down view.
(Direktlink, via Nag on the lake)
