Research map / 3D vision / 3D generation
Learning 3D Occupancy from Beam Overlap in 2D Rotating mmWave Radar
Yu Du, Ruifeng Nie, Long Ma, Chengpei Xu, Yu Liu, Weimin Wang
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 |
|---|---|---|---|
| use proposed two-stream beam-split network instead of U-Net | Radar occupancy network · Boreas | IoU: 15.19 -> 18.33 | Improved |
| use proposed network instead of U-Net | Radar occupancy network · ORR | IoU: 25.68 -> 28.60 | Improved |
| proposed network vs Occ4cast on ORR AP | Radar occupancy network · ORR | AP: 40.73 -> 40.64 | No clear effect |
| proposed network vs PCF/Occ4cast precision | Radar occupancy network · Boreas | Precision: 35.63 -> 23.24 | Hurt |
| split scan into two interleaved sweeps with 3 encoders (vs raw, 1 encoder) (ablation) | Radar occupancy network · Boreas | IoU: 17.98 -> 18.33 | Improved |
| split scan into two interleaved sweeps with 3 encoders (vs raw, 1 encoder) (ablation) | Radar occupancy network · ORR | IoU: 28.42 -> 28.60 | Improved |
| use two encoders on raw scan instead of one (ablation) | Radar occupancy network · ORR | IoU: 28.42 -> 28.26 | No clear effect |
| add BEV occupancy supervision and guidance (ablation) | Radar occupancy network · Boreas | IoU: 17.37 -> 18.33 | Improved |
| add BEV guidance path (fuse BEV probability with BEV features) (ablation) | Radar occupancy network · Boreas | IoU: 17.46 -> 18.33 | Improved |
| use polar 3D PE instead of Cartesian 2D PE (ablation) | Radar occupancy network · Boreas | IoU: 17.56 -> 18.33 | Improved |
| add elevation (z) PE to polar PE (ablation) | Radar occupancy network · Boreas | IoU: 18.23 -> 18.33 | No clear effect |
Closest papers by meaning
- RadarOcc: Robust 3D Occupancy Prediction with 4D Imaging Radar (NeurIPS 2024 · similarity 0.87)
- 3D Point Splatting for mmWave Radar Novel View Synthesis (NeurIPS 2026 · similarity 0.85)
- V2X-Radar: A Multi-modal Dataset with 4D Radar for Cooperative Perception (NeurIPS 2025 · similarity 0.83)
- Towards Robust 3D Object Detection with LiDAR and 4D Radar Fusion in Various Weather Conditions (CVPR 2024 · similarity 0.83)
- MITO: A Millimeter-Wave Dataset and Simulator for Non-Line-of-Sight Perception (NeurIPS 2026 · similarity 0.83)
- RC-AutoCalib: An End-to-End Radar-Camera Automatic Calibration Network (CVPR 2025 · similarity 0.82)
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.
