Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2511.21105 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912964119887872 |
|---|---|
| author | Mishra, Pushkal Bansal, Kshitiz Bharadia, Dinesh |
| author_facet | Mishra, Pushkal Bansal, Kshitiz Bharadia, Dinesh |
| contents | Radar sensors provide reliable perception across adverse weather, lighting, and long-range conditions, yet existing machine learning approaches remain fragmented and task-specific, with each downstream task employing distinct architectures and training objectives. We present RadarVLM, a vision-language framework that learns unified scene-level representations through structured spatial language supervision. Leveraging the CARLA simulator with a realistic radar model, we collect over 800k radar-caption pairs across 110+ hours of simulated driving in diverse scenarios. We make two key contributions: (1) a structured caption framework encoding vehicle distributions in the radar's native coordinate system, and (2) Spatially-Grounded CLIP (SG-CLIP) objective that replaces binary matching with continuous scene similarity, enabling fine-grained spatial reasoning. We further propose localization-aware evaluation metrics that directly assess spatial accuracy beyond traditional linguistic similarity measures. Validated on generative captioning and vehicle segmentation, SG-CLIP achieves up to 50% relative F1-score improvement over vanilla CLIP and a 21% AP gain on segmentation, demonstrating that language grounding produces spatially structured representations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_21105 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | RLM: A Vision-Language Model Approach for Radar Scene Understanding Mishra, Pushkal Bansal, Kshitiz Bharadia, Dinesh Computer Vision and Pattern Recognition Radar sensors provide reliable perception across adverse weather, lighting, and long-range conditions, yet existing machine learning approaches remain fragmented and task-specific, with each downstream task employing distinct architectures and training objectives. We present RadarVLM, a vision-language framework that learns unified scene-level representations through structured spatial language supervision. Leveraging the CARLA simulator with a realistic radar model, we collect over 800k radar-caption pairs across 110+ hours of simulated driving in diverse scenarios. We make two key contributions: (1) a structured caption framework encoding vehicle distributions in the radar's native coordinate system, and (2) Spatially-Grounded CLIP (SG-CLIP) objective that replaces binary matching with continuous scene similarity, enabling fine-grained spatial reasoning. We further propose localization-aware evaluation metrics that directly assess spatial accuracy beyond traditional linguistic similarity measures. Validated on generative captioning and vehicle segmentation, SG-CLIP achieves up to 50% relative F1-score improvement over vanilla CLIP and a 21% AP gain on segmentation, demonstrating that language grounding produces spatially structured representations. |
| title | RLM: A Vision-Language Model Approach for Radar Scene Understanding |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2511.21105 |