Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2603.07920 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912954515980288 |
|---|---|
| author | Qi, Zhangshuo Xu, Jingyi Cheng, Luqi Wen, Shichen Xiong, Guangming |
| author_facet | Qi, Zhangshuo Xu, Jingyi Cheng, Luqi Wen, Shichen Xiong, Guangming |
| contents | All-weather autonomy is critical for autonomous driving, which necessitates reliable localization across diverse scenarios. While LiDAR place recognition is widely deployed for this task, its performance degrades in adverse weather. Conversely, radar-based methods, though weather-resilient, are hindered by the general unavailability of radar maps. To bridge this gap, radar-to-LiDAR place recognition, which localizes radar scans within existing LiDAR maps, has garnered increasing interest. However, extracting discriminative and generalizable features shared between modalities remains challenging, compounded by the scarcity of large-scale paired training data and the signal heterogeneity across radar types. In this work, we propose RLPR, a robust radar-to-LiDAR place recognition framework compatible with single-chip, scanning, and 4D radars. We first design a dual-stream network to extract structural features that abstract away from sensor-specific signal properties (e.g., Doppler or RCS). Subsequently, motivated by our task-specific asymmetry observation between radar and LiDAR, we introduce a two-stage asymmetric cross-modal alignment (TACMA) strategy, which leverages the pre-trained radar branch as a discriminative anchor to guide the alignment process. Experiments on four datasets demonstrate that RLPR achieves state-of-the-art recognition accuracy with strong zero-shot generalization capabilities. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_07920 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | RLPR: Radar-to-LiDAR Place Recognition via Two-Stage Asymmetric Cross-Modal Alignment for Autonomous Driving Qi, Zhangshuo Xu, Jingyi Cheng, Luqi Wen, Shichen Xiong, Guangming Computer Vision and Pattern Recognition All-weather autonomy is critical for autonomous driving, which necessitates reliable localization across diverse scenarios. While LiDAR place recognition is widely deployed for this task, its performance degrades in adverse weather. Conversely, radar-based methods, though weather-resilient, are hindered by the general unavailability of radar maps. To bridge this gap, radar-to-LiDAR place recognition, which localizes radar scans within existing LiDAR maps, has garnered increasing interest. However, extracting discriminative and generalizable features shared between modalities remains challenging, compounded by the scarcity of large-scale paired training data and the signal heterogeneity across radar types. In this work, we propose RLPR, a robust radar-to-LiDAR place recognition framework compatible with single-chip, scanning, and 4D radars. We first design a dual-stream network to extract structural features that abstract away from sensor-specific signal properties (e.g., Doppler or RCS). Subsequently, motivated by our task-specific asymmetry observation between radar and LiDAR, we introduce a two-stage asymmetric cross-modal alignment (TACMA) strategy, which leverages the pre-trained radar branch as a discriminative anchor to guide the alignment process. Experiments on four datasets demonstrate that RLPR achieves state-of-the-art recognition accuracy with strong zero-shot generalization capabilities. |
| title | RLPR: Radar-to-LiDAR Place Recognition via Two-Stage Asymmetric Cross-Modal Alignment for Autonomous Driving |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.07920 |