DeepCrashTest: Turning Dashcam Videos into Virtual Crash Tests for Automated Driving Systems
DeepCrashTest: Turning Dashcam Videos into Virtual Crash Tests for Automated Driving Systems
复制标题
DOI：
10.1109/icra40945.2020.9197053
复制
发表时间：
2020-03
期刊：
影响因子：
--
通讯作者：
Sai Krishna Bashetty;H. B. Amor;Georgios Fainekos
中科院分区：
文献类型：
--
作者：
Sai Krishna Bashetty;H. B. Amor;Georgios Fainekos
本文的目标是生成具有真实碰撞场景的模拟，用于训练和测试自动驾驶车辆。我们使用互联网上上传的大量dashcam碰撞视频来提取有价值的碰撞数据，并在模拟器中重现碰撞场景。我们解决的问题，提取3D车辆轨迹从视频记录的一个未知的和未校准的单目摄像机源使用模块化的方法。一个工作架构和演示视频沿着与开放源代码的实现提供的文件。
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.