DeepCrashTest: Turning Dashcam Videos into Virtual Crash Tests for Automated Driving Systems
This provides a method to create virtual crash tests for automated driving systems, addressing a domain-specific need for training and testing.
The paper tackles the problem of generating realistic crash simulations for autonomous vehicles by extracting 3D vehicle trajectories from uncalibrated dashcam videos, resulting in a working architecture and open-source implementation.
The goal of this paper is to generate simulations with real-world collision scenarios for training and testing autonomous vehicles. We use numerous dashcam crash videos uploaded on the internet to extract valuable collision data and recreate the crash scenarios in a simulator. We tackle the problem of extracting 3D vehicle trajectories from videos recorded by an unknown and uncalibrated monocular camera source using a modular approach. A working architecture and demonstration videos along with the open-source implementation are provided with the paper.