Synthesizing Data for Autonomous Driving: Multi-Agent Reinforcement
Learning Meets Augmented Reality
Author:
Meng Chao1, Zhang Song2, Wang Hanchao1, Gu Kai1, Wang Tong1, Mei Jinren1
Affiliation:
1. Z-one Technology co., Ltd. 2. Shanghai Automotive Industry Corporation: SAIC Motor Corpora
Abstract
<div class="section abstract"><div class="htmlview paragraph">Synthetic data holds significant potential for improving the efficiency of
perception tasks in autonomous driving. This paper proposes a practical data
synthesis pipeline that employs multi-agent reinforcement learning (MARL) to
automatically generate dynamic traffic participant trajectories and leverages
augmented reality (AR) processes to produce photo-realistic images. This AR
process blends clean static background images extracted from real photos using
image matting techniques, with dynamic foreground images rendered from 3D
Computer Aided Design (CAD) models in a rendering engine. We posit that this
data synthetic pipe line has strong image photorealism, flexible way of
interaction scenarios generation and mature tool chain, which has the prospect
of large-scale engineering application.</div></div>
Publisher
SAE International
Reference39 articles.
1. Alhaija ,
H. ,
Mustikovela ,
S.K. ,
Mescheder ,
L.M. ,
Geiger ,
A.
et al. Augmented Reality Meets
Computer Vision: Efficient Data Generation for Urban Driving
Scenes Int. J. Comput. Vis. 126 9 2018 961 972 2. Bojarski ,
M. ,
Del Testa ,
D. ,
Dworakowski ,
D. ,
Firner ,
B.
et al. End to End Learning for
Self-Driving Cars 2016 3. Brockman ,
G. ,
Cheung ,
V. ,
Pettersson ,
L. ,
Schneider ,
J.
et al. OpenAI
Gym 2016 4. Chen ,
B. ,
Chen ,
X. ,
Qiong , W. ,
and
Li ,
L. Adversarial Evaluation of Autonomous Vehicles in Lane-Change
Scenarios IEEE Trans. Intell. Transp.
Syst. 23 8 2022 10333 10342 5. Chen ,
L.-C. ,
Papandreou ,
G. ,
Schroff , F. ,
and
Adam ,
H. 2017
|
|