Research map / Detection and segmentation / 3D detection and driving
Doppler-Aware LiDAR-RADAR Fusion for Weather-Robust 3D Detection
Yujeong Chae, Heejun Park, Hyeonseong Kim, Kuk-Jin Yoon
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 |
|---|---|---|---|
| add Doppler as fifth input dimension to power (ablation) | RADAR-only RTNH · K-RADAR Total, IoU=0.5 (RADAR-only model) | AP_3D: 14.1 -> 15.7 | Improved |
| add Doppler motion-compensation preprocessing (Doppler only) (ablation) | RADAR-only RTNH · K-RADAR Total, IoU=0.5 (RADAR-only model) | AP_3D: 14.6 -> 18.7 | Improved |
| add Doppler motion-compensation preprocessing (power + Doppler) (ablation) | RADAR-only RTNH · K-RADAR Total, IoU=0.5 (RADAR-only model) | AP_3D: 15.7 -> 16.3 | Improved |
| encode power and Doppler in separate branches (ablation) | RADAR-only RTNH · K-RADAR Total, IoU=0.5 (RADAR-only model) | AP_3D: 16.3 -> 19.9 | Improved |
| encode power and Doppler in separate branches (ablation) | RADAR-only RTNH · K-RADAR Total, IoU=0.5 (RADAR-only model) | AP_BEV: 45.4 -> 49.9 | Improved |
| add Doppler-LiDAR interaction to power-LiDAR (ablation) | DLRFusion · K-RADAR Total, IoU=0.5 (fusion) | AP_3D: 37.9 -> 37.9 | No clear effect |
| add power-Doppler interaction to P-L and D-L (ablation) | DLRFusion · K-RADAR Total, IoU=0.5 (fusion) | AP_3D: 37.9 -> 45.7 | Improved |
| add power-Doppler interaction to P-L and D-L (ablation) | DLRFusion · K-RADAR Total, IoU=0.5 (fusion) | AP_BEV: 71.2 -> 73.2 | Improved |
| use multi-path iterative interaction instead of concatenation (ablation) | DLRFusion · K-RADAR Total, IoU=0.5 (fusion) | AP_3D: 36.9 -> 45.7 | Improved |
| use multi-path iterative interaction instead of one-way voxel attention (ablation) | DLRFusion · K-RADAR Total, IoU=0.5 (fusion) | AP_3D: 37.9 -> 45.7 | Improved |
| use multi-path iterative interaction instead of BEV-space fusion (ablation) | DLRFusion · K-RADAR Total, IoU=0.5 (fusion) | AP_3D: 31.0 -> 45.7 | Improved |
| Doppler-aware fusion versus LOD PDR on total split (ablation) | DLRFusion · K-RADAR Total, IoU=0.5 | AP_3D: 40.4 -> 45.7 | Improved |
Closest papers by meaning
- Towards Robust 3D Object Detection with LiDAR and 4D Radar Fusion in Various Weather Conditions (CVPR 2024 · similarity 0.93)
- L4DR: LiDAR-4DRadar Fusion for Weather-Robust 3D Object Detection (AAAI 2025 · similarity 0.92)
- RadarOcc: Robust 3D Occupancy Prediction with 4D Imaging Radar (NeurIPS 2024 · similarity 0.88)
- RobuRCDet: Enhancing Robustness of Radar-Camera Fusion in Bird's Eye View for 3D Object Detection (ICLR 2025 · similarity 0.88)
- RadarDistill: Boosting Radar-based Object Detection Performance via Knowledge Distillation from LiDAR Features (CVPR 2024 · similarity 0.87)
- Sunshine to Rainstorm: Cross-Weather Knowledge Distillation for Robust 3D Object Detection (AAAI 2024 · similarity 0.87)
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.
