Print Email Facebook Twitter Hidden Gems: 4D Radar Scene Flow Learning Using Cross-Modal Supervision Title Hidden Gems: 4D Radar Scene Flow Learning Using Cross-Modal Supervision Author Ding, Fangqiang (University of Edinburgh) Palffy, A. (TU Delft Microwave Sensing, Signals & Systems; TU Delft Intelligent Vehicles) Gavrila, D. (TU Delft Intelligent Vehicles) Lu, Chris Xiaoxuan (University of Edinburgh) Contributor O'Conner, Lisa (editor) Date 2023 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. To reference this document use: http://resolver.tudelft.nl/uuid:ac62ef7b-30ab-4a28-85c1-bb4fc0ef6672 DOI https://doi.org/10.1109/CVPR52729.2023.00901 Publisher IEEE, Piscataway Embargo date 2024-02-22 ISBN 979-8-3503-0130-4 Source Proceedings of the 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Event 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023-06-17 → 2023-06-24, Vancouver, Canada 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. Part of collection Institutional Repository Document type conference paper Rights © 2023 Fangqiang Ding, A. Palffy, D. Gavrila, Chris Xiaoxuan Lu Files PDF Hidden_Gems_4D_Radar_Scen ... vision.pdf 1.33 MB Close viewer /islandora/object/uuid:ac62ef7b-30ab-4a28-85c1-bb4fc0ef6672/datastream/OBJ/view