Research map / Detection and segmentation / 3D detection and driving
R-LiViT: A LiDAR-Visual-Thermal Dataset Enabling Vulnerable Road User Focused Roadside Perception
Jonas Mirlach, Lei Wan, Andreas Wiedholz, Hannan Ejaz Keen, Andreas Eich
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| train on anonymized RGB images instead of originals (ablation) | Faster R-CNN · R-LiViT RGB test | AP_per: 37.65 -> 37.55 | No clear effect |
| train on anonymized RGB images instead of originals (ablation) | YOLOv8 · R-LiViT RGB test | AP_veh: 34.92 -> 35.25 | No clear effect |
| train on anonymized RGB images instead of originals (ablation) | RT-DETR · R-LiViT RGB test | AP_per: 31.88 -> 31.29 | No clear effect |
| add thermal via late fusion to RGB detector at night (ablation) | YOLOv8 · R-LiViT RGBT (night) | AP: 32.27 -> 33.95 | Improved |
| add thermal via late fusion to RGB detector at night (ablation) | RT-DETR · R-LiViT RGBT (night) | AP: 32.67 -> 34.46 | Improved |
| add thermal via late fusion to RGB detector at night (ablation) | Faster R-CNN · R-LiViT RGBT (night) | AP: 24.98 -> 20.47 | Hurt |
| add thermal via late fusion to RGB detector in daytime (ablation) | YOLOv8 · R-LiViT RGBT (day) | AP: 44.87 -> 43.09 | Hurt |
| add thermal late fusion to RGB detector (ablation) | YOLOv8 · R-LiViT RGBT (day) | AR100 person: 46.39 -> 50.42 | Improved |
| use thermal instead of RGB input (ablation) | YOLOv8 · R-LiViT RGBT (day) | AP: 44.87 -> 31.90 | Hurt |
| use point-based PointRCNN instead of PointPillars (ablation) | LiDAR 3D detection · R-LiViT (Car) | AP75: 46.73 -> 28.26 | Hurt |
| use AB3DMOT instead of SimpleTrack (ablation) | 3D MOT on PointPillars detections · R-LiViT (Car) | sAMOTA: 64.40 -> 78.45 | Improved |
Closest papers by meaning
- RCooper: A Real-world Large-scale Dataset for Roadside Cooperative Perception (CVPR 2024 · similarity 0.87)
- LidaRF: Delving into Lidar for Neural Radiance Field on Street Scenes (CVPR 2024 · similarity 0.87)
- Effective Multi-sensor Conditioning for Street-view Novel-view Synthesis (NeurIPS 2026 · similarity 0.86)
- V2X-Radar: A Multi-modal Dataset with 4D Radar for Cooperative Perception (NeurIPS 2025 · similarity 0.85)
- UrbanIng-V2X: A Large-Scale Multi-Vehicle, Multi-Infrastructure Dataset Across Multiple Intersections for Cooperative Perception (NeurIPS 2025 · similarity 0.85)
- The Road Ahead in Autonomous Driving: The KITScenes Multimodal Dataset (NeurIPS 2026 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
