DoGFlow: Self-Supervised LiDAR Scene Flow via Cross-Modal Doppler Guidance

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Khoche, Ajinkya, Zhang, Qingwen, Cai, Yixi, Mansouri, Sina Sharif, Jensfelt, Patric
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918130581766144
author Khoche, Ajinkya
Zhang, Qingwen
Cai, Yixi
Mansouri, Sina Sharif
Jensfelt, Patric
author_facet Khoche, Ajinkya
Zhang, Qingwen
Cai, Yixi
Mansouri, Sina Sharif
Jensfelt, Patric
contents Accurate 3D scene flow estimation is critical for autonomous systems to navigate dynamic environments safely, but creating the necessary large-scale, manually annotated datasets remains a significant bottleneck for developing robust perception models. Current self-supervised methods struggle to match the performance of fully supervised approaches, especially in challenging long-range and adverse weather scenarios, while supervised methods are not scalable due to their reliance on expensive human labeling. We introduce DoGFlow, a novel self-supervised framework that recovers full 3D object motions for LiDAR scene flow estimation without requiring any manual ground truth annotations. This paper presents our cross-modal label transfer approach, where DoGFlow computes motion pseudo-labels in real-time directly from 4D radar Doppler measurements and transfers them to the LiDAR domain using dynamic-aware association and ambiguity-resolved propagation. On the challenging MAN TruckScenes dataset, DoGFlow substantially outperforms existing self-supervised methods and improves label efficiency by enabling LiDAR backbones to achieve over 90% of fully supervised performance with only 10% of the ground truth data. For more details, please visit https://ajinkyakhoche.github.io/DogFlow/
format Preprint
id arxiv_https___arxiv_org_abs_2508_18506
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DoGFlow: Self-Supervised LiDAR Scene Flow via Cross-Modal Doppler Guidance
Khoche, Ajinkya
Zhang, Qingwen
Cai, Yixi
Mansouri, Sina Sharif
Jensfelt, Patric
Computer Vision and Pattern Recognition
Accurate 3D scene flow estimation is critical for autonomous systems to navigate dynamic environments safely, but creating the necessary large-scale, manually annotated datasets remains a significant bottleneck for developing robust perception models. Current self-supervised methods struggle to match the performance of fully supervised approaches, especially in challenging long-range and adverse weather scenarios, while supervised methods are not scalable due to their reliance on expensive human labeling. We introduce DoGFlow, a novel self-supervised framework that recovers full 3D object motions for LiDAR scene flow estimation without requiring any manual ground truth annotations. This paper presents our cross-modal label transfer approach, where DoGFlow computes motion pseudo-labels in real-time directly from 4D radar Doppler measurements and transfers them to the LiDAR domain using dynamic-aware association and ambiguity-resolved propagation. On the challenging MAN TruckScenes dataset, DoGFlow substantially outperforms existing self-supervised methods and improves label efficiency by enabling LiDAR backbones to achieve over 90% of fully supervised performance with only 10% of the ground truth data. For more details, please visit https://ajinkyakhoche.github.io/DogFlow/
title DoGFlow: Self-Supervised LiDAR Scene Flow via Cross-Modal Doppler Guidance
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2508.18506