NeuRAD: Neural Rendering for Autonomous Driving
(2024) 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) p.14895-14904- Abstract
- Neural radiance fields (NeRFs) have gained popularity in the autonomous driving (AD) community. Recent meth-ods show NeRFs' potential for closed-loop simulation, en-abling testing of AD systems, and as an advanced training data augmentation technique. However, existing meth-ods often require long training times, dense semantic su-pervision, or lack generalizability. This, in turn, hinders the application of NeRFs for AD at scale. In this paper, we propose NeuRAD, a robust novel view synthesis method tailored to dynamic AD data. Our method features simple network design, extensive sensor modeling for both cam-era and lidar - including rolling shutter, beam divergence and ray dropping - and is applicable to multiple datasets out of the box.... (More)
- Neural radiance fields (NeRFs) have gained popularity in the autonomous driving (AD) community. Recent meth-ods show NeRFs' potential for closed-loop simulation, en-abling testing of AD systems, and as an advanced training data augmentation technique. However, existing meth-ods often require long training times, dense semantic su-pervision, or lack generalizability. This, in turn, hinders the application of NeRFs for AD at scale. In this paper, we propose NeuRAD, a robust novel view synthesis method tailored to dynamic AD data. Our method features simple network design, extensive sensor modeling for both cam-era and lidar - including rolling shutter, beam divergence and ray dropping - and is applicable to multiple datasets out of the box. We verify its performance on five popular AD datasets, achieving state-of-the-art performance across the board. To encourage further development, we openly release the NeuRAD source code. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/dcda7743-11f9-4720-a0a5-110371f106c9
- author
- Tonderski, Adam
LU
; Lindström, Carl ; Hess, Georg ; Ljungbergh, William ; Svensson, Lennart and Petersson, Christoffer
- organization
- publishing date
- 2024-06-16
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- host publication
- 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
- pages
- 14895 - 14904
- publisher
- IEEE - Institute of Electrical and Electronics Engineers Inc.
- conference name
- 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
- conference dates
- 2024-06-16 - 2024-06-22
- external identifiers
-
- scopus:85190896037
- ISBN
- 979-8-3503-5300-6
- DOI
- 10.1109/CVPR52733.2024.01411
- language
- English
- LU publication?
- yes
- id
- dcda7743-11f9-4720-a0a5-110371f106c9
- date added to LUP
- 2025-01-31 10:23:59
- date last changed
- 2025-04-04 14:23:42
@inproceedings{dcda7743-11f9-4720-a0a5-110371f106c9, abstract = {{Neural radiance fields (NeRFs) have gained popularity in the autonomous driving (AD) community. Recent meth-ods show NeRFs' potential for closed-loop simulation, en-abling testing of AD systems, and as an advanced training data augmentation technique. However, existing meth-ods often require long training times, dense semantic su-pervision, or lack generalizability. This, in turn, hinders the application of NeRFs for AD at scale. In this paper, we propose NeuRAD, a robust novel view synthesis method tailored to dynamic AD data. Our method features simple network design, extensive sensor modeling for both cam-era and lidar - including rolling shutter, beam divergence and ray dropping - and is applicable to multiple datasets out of the box. We verify its performance on five popular AD datasets, achieving state-of-the-art performance across the board. To encourage further development, we openly release the NeuRAD source code.}}, author = {{Tonderski, Adam and Lindström, Carl and Hess, Georg and Ljungbergh, William and Svensson, Lennart and Petersson, Christoffer}}, booktitle = {{2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)}}, isbn = {{979-8-3503-5300-6}}, language = {{eng}}, month = {{06}}, pages = {{14895--14904}}, publisher = {{IEEE - Institute of Electrical and Electronics Engineers Inc.}}, title = {{NeuRAD: Neural Rendering for Autonomous Driving}}, url = {{http://dx.doi.org/10.1109/CVPR52733.2024.01411}}, doi = {{10.1109/CVPR52733.2024.01411}}, year = {{2024}}, }