Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Super4DR: 4D Radar-centric Self-supervised Odometry and Gaussian-based Map Optimization

About

Conventional SLAM systems using visual or LiDAR data often struggle in poor lighting and severe weather. Although 4D radar is suited for such environments, its sparse and noisy point clouds hinder accurate odometry estimation, while the radar maps suffer from obscure and incomplete structures. Thus, we propose Super4DR, a 4D radar-centric framework for learning-based odometry estimation and gaussian-based map optimization. First, we design a cluster-aware odometry network that incorporates object-level cues from the clustered radar points for inter-frame matching, alongside a hierarchical self-supervision mechanism to overcome outliers through spatio-temporal consistency, knowledge transfer, and feature contrast. Second, we propose using 3D gaussians as an intermediate representation, coupled with a radar-specific growth strategy, selective separation, and multi-view regularization, to recover blurry map areas and those undetected based on image texture. Experiments show that Super4DR achieves a 67% performance gain over prior self-supervised methods, nearly matches supervised odometry, and narrows the map quality disparity with LiDAR while enabling multi-modal image rendering.

Zhiheng Li, Weihua Wang, Qiang Shen, Yichen Zhao, Zheng Fang• 2025

Related benchmarks

TaskDatasetResultRank
OdometryView-of-Delft (VoD) Mean
t_rel (Translation Error)0.05
14
OdometryView-of-Delft (VoD) sequence 09
t_rel (Translation Error)0.04
14
OdometryView-of-Delft (VoD) sequence 17
t_rel (Translation Error)0.03
14
OdometryView-of-Delft (VoD) sequence 19
t_rel (Translation Error)0.04
14
OdometryView-of-Delft (VoD) sequence 22
t_rel Error0.06
14
OdometryView-of-Delft (VoD) sequence 24
t_rel0.12
14
OdometryView-of-Delft (VoD) sequence 04
Rel. Translation Error (t_rel)4
14
OdometryView-of-Delft (VoD) sequence 03
Rel. Translation Error (t_rel)0.04
12
Image RenderingNTU4DRadLM Garden (Unstructured)
PSNR25.875
8
4D Radar OdometryNTU4DRadLM 0.25km (Cp)
RPE translation (m)0.248
7
Showing 10 of 35 rows

Other info

Follow for update