Research map / Detection and segmentation / 3D detection and driving
SparseRadNet: Sparse Perception Neural Network on Subsampled Radar Data
Jialong Wu, Mirko Meuter, Markus Schoeler, Matthias Rottmann
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 |
|---|---|---|---|
| learned deep radar subsampling instead of CA-CFAR top-M SNR selection (ablation) | FFT-RadNet with 4000 samples · RADIal test | F1: 86.93 -> 90.26 | Improved |
| learned DRS instead of top-M energy selection (ablation) | FFT-RadNet with 1000 samples · RADIal test | F1: 82.15 -> 88.13 | Improved |
| add GNN branch to sparse SCNN branch (ablation) | SparseRadNet backbone · RADIal test | F1: 89.50 -> 91.71 | Improved |
| add axial attentive fusion to two-branch backbone (ablation) | SparseRadNet · RADIal test | F1: 91.71 -> 93.84 | Improved |
| use sparse SCNN-only backbone instead of dense backbone (ablation) | DRS + backbone · RADIal test | F1: 90.26 -> 89.50 | Hurt |
| subsample input to 4000 pixels (3% of RD spectrum) with DRS (ablation) | FFT-RadNet · RADIal test | F1: 90.75 -> 90.26 | No clear effect |
| SparseRadNet versus TransRadar on hard cases | SparseRadNet · RADIal hard subset | F1: 85.68 -> 85.45 | Hurt |
| sparse subsampled model on dense freespace segmentation | SparseRadNet · RADIal freespace segmentation | mIoU: TransRadar 82.27 vs SparseRadNet 78.48 | Mixed |
Closest papers by meaning
- AdaRadar: Rate Adaptive Spectral Compression for Radar-based Perception (CVPR 2026 · similarity 0.87)
- NEAR: Neural Electromagnetic Array Response (ICML 2025 · similarity 0.85)
- V2X-Radar: A Multi-modal Dataset with 4D Radar for Cooperative Perception (NeurIPS 2025 · similarity 0.84)
- Bootstrapping Autonomous Driving Radars with Self-Supervised Learning (CVPR 2024 · similarity 0.84)
- AdaPKC: PeakConv with Adaptive Peak Receptive Field for Radar Semantic Segmentation (NeurIPS 2024 · similarity 0.84)
- SparseAlign: a Fully Sparse Framework for Cooperative Object Detection (CVPR 2025 · similarity 0.84)
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.
