Hidden Gems: 4D Radar Scene Flow Learning Using Cross-Modal Supervision

Conference Paper (2023)
Author(s)

Fangqiang Ding (The University of Edinburgh)

Andras Palffy (TU Delft - Intelligent Vehicles, TU Delft - Microwave Sensing, Signals & Systems)

D. M. Gavrila (TU Delft - Intelligent Vehicles)

Chris Xiaoxuan Lu (The University of Edinburgh)

Microwave Sensing, Signals & Systems
Copyright
© 2023 Fangqiang Ding, A. Palffy, D. Gavrila, Chris Xiaoxuan Lu
DOI related publication
https://doi.org/10.1109/CVPR52729.2023.00901
More Info
expand_more
Publication Year
2023
Language
English
Copyright
© 2023 Fangqiang Ding, A. Palffy, D. Gavrila, Chris Xiaoxuan Lu
Microwave Sensing, Signals & Systems
Bibliographical Note
Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.@en
Volume number
2023-June
Pages (from-to)
9340-9349
ISBN (print)
979-8-3503-0130-4
ISBN (electronic)
979-8-3503-0129-8
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

This work proposes a novel approach to 4D radar-based scene flow estimation via cross-modal learning. Our approach is motivated by the co-located sensing redundancy in modern autonomous vehicles. Such redundancy implicitly provides various forms of supervision cues to the radar scene flow estimation. Specifically, we introduce a multi-task model architecture for the identified cross-modal learning problem and propose loss functions to opportunistically engage scene flow estimation using multiple cross-modal constraints for effective model training. Extensive experiments show the state-of-the-art performance of our method and demonstrate the effectiveness of cross-modal super-vised learning to infer more accurate 4D radar scene flow. We also show its usefulness to two subtasks - motion segmentation and ego-motion estimation. Our source code will be available on https://github.com/Toytiny/CMFlow.

Files

Hidden_Gems_4D_Radar_Scene_Flo... (pdf)
(pdf | 1.33 Mb)
- Embargo expired in 22-02-2024
License info not available