DNOI-4DRO: Deep 4D Radar Odometry with Differentiable Neural-Optimization Iterations

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Lu, Shouyi, Zhou, Huanyu, Zhuo, Guirong, Tang, Xiao
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866908637104963584
author Lu, Shouyi
Zhou, Huanyu
Zhuo, Guirong
Tang, Xiao
author_facet Lu, Shouyi
Zhou, Huanyu
Zhuo, Guirong
Tang, Xiao
contents A novel learning-optimization-combined 4D radar odometry model, named DNOI-4DRO, is proposed in this paper. The proposed model seamlessly integrates traditional geometric optimization with end-to-end neural network training, leveraging an innovative differentiable neural-optimization iteration operator. In this framework, point-wise motion flow is first estimated using a neural network, followed by the construction of a cost function based on the relationship between point motion and pose in 3D space. The radar pose is then refined using Gauss-Newton updates. Additionally, we design a dual-stream 4D radar backbone that integrates multi-scale geometric features and clustering-based class-aware features to enhance the representation of sparse 4D radar point clouds. Extensive experiments on the VoD and Snail-Radar datasets demonstrate the superior performance of our model, which outperforms recent classical and learning-based approaches. Notably, our method even achieves results comparable to A-LOAM with mapping optimization using LiDAR point clouds as input. Our models and code will be publicly released.
format Preprint
id arxiv_https___arxiv_org_abs_2505_12310
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DNOI-4DRO: Deep 4D Radar Odometry with Differentiable Neural-Optimization Iterations
Lu, Shouyi
Zhou, Huanyu
Zhuo, Guirong
Tang, Xiao
Computer Vision and Pattern Recognition
Artificial Intelligence
Robotics
A novel learning-optimization-combined 4D radar odometry model, named DNOI-4DRO, is proposed in this paper. The proposed model seamlessly integrates traditional geometric optimization with end-to-end neural network training, leveraging an innovative differentiable neural-optimization iteration operator. In this framework, point-wise motion flow is first estimated using a neural network, followed by the construction of a cost function based on the relationship between point motion and pose in 3D space. The radar pose is then refined using Gauss-Newton updates. Additionally, we design a dual-stream 4D radar backbone that integrates multi-scale geometric features and clustering-based class-aware features to enhance the representation of sparse 4D radar point clouds. Extensive experiments on the VoD and Snail-Radar datasets demonstrate the superior performance of our model, which outperforms recent classical and learning-based approaches. Notably, our method even achieves results comparable to A-LOAM with mapping optimization using LiDAR point clouds as input. Our models and code will be publicly released.
title DNOI-4DRO: Deep 4D Radar Odometry with Differentiable Neural-Optimization Iterations
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Robotics
url https://arxiv.org/abs/2505.12310