Research map / 3D vision / Depth and stereo
Raw High-Definition Radar for Multi-Task Learning
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 RD input with learned angle recovery instead of RA input with conventional processing | FFT-RadNet vs Pixor-RA · RADIal Test split, Overall | AP (%): 96.56 -> 96.84 | Helped |
| Use RD input with learned angle recovery instead of RA input with conventional processing | FFT-RadNet vs Pixor-RA · RADIal Test split, Overall | AR (%): 81.68 -> 82.18 | Helped |
| Use RD input with learned angle recovery instead of RA input with conventional processing | FFT-RadNet vs Pixor-RA · RADIal Test split, Overall | Angle error (°): 0.20 -> 0.17 | Helped |
| Use RD input with learned angle recovery instead of RA input with conventional processing | FFT-RadNet vs Pixor-RA · RADIal Test split, Overall | Range error (m): 0.10 -> 0.11 | Hurt |
| Use RD input with learned angle recovery instead of RA input with conventional processing | FFT-RadNet vs Pixor-RA · RADIal Test split, Easy | AP (%): +1.6% AP | Helped |
| Use RD input with learned angle recovery instead of RA input with conventional processing | FFT-RadNet vs Pixor-RA · RADIal Test split, Easy | AR (%): +3.6% AR | Helped |
| Use RD input with learned angle recovery instead of RA input with conventional processing | FFT-RadNet vs Pixor-RA · RADIal Test split, Hard | AR (%): 70.10 -> 64.82 | Hurt |
| Use RD input with learned angle recovery instead of RA input | FFT-RadNet vs PolarNet · RADIal free-driving-space segmentation, Overall | mIoU (%): 60.6 -> 74.0; 13.4% IoU | Helped |
| Use RD input instead of point-cloud input | FFT-RadNet vs Pixor-PC · RADIal Test split, Overall | AR (%): 32.32 -> 82.18 | Helped |
| Reduce pre-encoder output channels to 192 rather than using the maximum 384 (ablation) | FFT-RadNet MIMO pre-encoder · RADIal validation set | Detection F1-score | Helped |
| Compress the pre-encoder output to fewer channels (ablation) | FFT-RadNet MIMO pre-encoder · RADIal validation set | Detection F1-score | Mixed |
Closest papers by meaning
- RadSimReal: Bridging the Gap Between Synthetic and Real Data in Radar Object Detection With Simulation (CVPR 2024 · similarity 0.84)
- RadarOcc: Robust 3D Occupancy Prediction with 4D Imaging Radar (NeurIPS 2024 · similarity 0.84)
- DART: Implicit Doppler Tomography for Radar Novel View Synthesis (CVPR 2024 · similarity 0.84)
- RadarMOSEVE: A Spatial-Temporal Transformer Network for Radar-Only Moving Object Segmentation and Ego-Velocity Estimation (AAAI 2024 · similarity 0.84)
- NEAR: Neural Electromagnetic Array Response (ICML 2025 · similarity 0.83)
- V2X-Radar: A Multi-modal Dataset with 4D Radar for Cooperative Perception (NeurIPS 2025 · similarity 0.83)
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.