MARS: An Instance-aware, Modular and Realistic Simulator for Autonomous Driving

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wu, Zirui, Liu, Tianyu, Luo, Liyi, Zhong, Zhide, Chen, Jianteng, Xiao, Hongmin, Hou, Chao, Lou, Haozhe, Chen, Yuantao, Yang, Runyi, Huang, Yuxin, Ye, Xiaoyu, Yan, Zike, Shi, Yongliang, Liao, Yiyi, Zhao, Hao
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929474992340992
author Wu, Zirui
Liu, Tianyu
Luo, Liyi
Zhong, Zhide
Chen, Jianteng
Xiao, Hongmin
Hou, Chao
Lou, Haozhe
Chen, Yuantao
Yang, Runyi
Huang, Yuxin
Ye, Xiaoyu
Yan, Zike
Shi, Yongliang
Liao, Yiyi
Zhao, Hao
author_facet Wu, Zirui
Liu, Tianyu
Luo, Liyi
Zhong, Zhide
Chen, Jianteng
Xiao, Hongmin
Hou, Chao
Lou, Haozhe
Chen, Yuantao
Yang, Runyi
Huang, Yuxin
Ye, Xiaoyu
Yan, Zike
Shi, Yongliang
Liao, Yiyi
Zhao, Hao
contents Nowadays, autonomous cars can drive smoothly in ordinary cases, and it is widely recognized that realistic sensor simulation will play a critical role in solving remaining corner cases by simulating them. To this end, we propose an autonomous driving simulator based upon neural radiance fields (NeRFs). Compared with existing works, ours has three notable features: (1) Instance-aware. Our simulator models the foreground instances and background environments separately with independent networks so that the static (e.g., size and appearance) and dynamic (e.g., trajectory) properties of instances can be controlled separately. (2) Modular. Our simulator allows flexible switching between different modern NeRF-related backbones, sampling strategies, input modalities, etc. We expect this modular design to boost academic progress and industrial deployment of NeRF-based autonomous driving simulation. (3) Realistic. Our simulator set new state-of-the-art photo-realism results given the best module selection. Our simulator will be open-sourced while most of our counterparts are not. Project page: https://open-air-sun.github.io/mars/.
format Preprint
id arxiv_https___arxiv_org_abs_2307_15058
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle MARS: An Instance-aware, Modular and Realistic Simulator for Autonomous Driving
Wu, Zirui
Liu, Tianyu
Luo, Liyi
Zhong, Zhide
Chen, Jianteng
Xiao, Hongmin
Hou, Chao
Lou, Haozhe
Chen, Yuantao
Yang, Runyi
Huang, Yuxin
Ye, Xiaoyu
Yan, Zike
Shi, Yongliang
Liao, Yiyi
Zhao, Hao
Computer Vision and Pattern Recognition
Nowadays, autonomous cars can drive smoothly in ordinary cases, and it is widely recognized that realistic sensor simulation will play a critical role in solving remaining corner cases by simulating them. To this end, we propose an autonomous driving simulator based upon neural radiance fields (NeRFs). Compared with existing works, ours has three notable features: (1) Instance-aware. Our simulator models the foreground instances and background environments separately with independent networks so that the static (e.g., size and appearance) and dynamic (e.g., trajectory) properties of instances can be controlled separately. (2) Modular. Our simulator allows flexible switching between different modern NeRF-related backbones, sampling strategies, input modalities, etc. We expect this modular design to boost academic progress and industrial deployment of NeRF-based autonomous driving simulation. (3) Realistic. Our simulator set new state-of-the-art photo-realism results given the best module selection. Our simulator will be open-sourced while most of our counterparts are not. Project page: https://open-air-sun.github.io/mars/.
title MARS: An Instance-aware, Modular and Realistic Simulator for Autonomous Driving
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2307.15058