Research map / Detection and segmentation / 3D detection and driving
Multi-View Radar Semantic Segmentation
Arthur Ouaknine, Alasdair Newson, Patrick Pérez, Florence Tupin, Julien Rebut
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 ASPP modules to MV-Net (ablation) | MV-Net · CARRADA-Test range-Doppler | mIoU (%): 29.0 -> 48.9 | Improved |
| add ASPP modules to MV-Net (ablation) | MV-Net · CARRADA-Test range-angle | mIoU (%): 26.8 -> 28.1 | Improved |
| add AD-view encoder to MVA-Net (a) (ablation) | MVA-Net (a) · CARRADA-Test range-Doppler | mIoU (%): 48.9 -> 52.9 | Improved |
| add AD-view encoder to MVA-Net (a) (ablation) | MVA-Net (a) · CARRADA-Test range-angle | mIoU (%): 28.1 -> 36.7 | Improved |
| add 3D temporal convolutions to MVA-Net (b) (ablation) | MVA-Net (b) · CARRADA-Test range-Doppler | mIoU (%): 52.9 -> 59.3 | Improved |
| add 3D temporal convolutions to MVA-Net (b) (ablation) | MVA-Net (b) · CARRADA-Test range-angle | mIoU (%): 36.7 -> 40.1 | Improved |
| add coherence loss (CoL) to wCE+SDice (ablation) | TMVA-Net · CARRADA-Test range-angle | mIoU (%): 40.1 -> 41.3 | Improved |
| add coherence loss (CoL) to wCE+SDice (ablation) | TMVA-Net · CARRADA-Test range-Doppler | mIoU (%): 59.3 -> 58.7 | Hurt |
| use class-weighted CE instead of CE in CE+SDice (ablation) | TMVA-Net · CARRADA-Test range-Doppler | mIoU (%): 45.2 -> 59.3 | Improved |
| add Soft Dice to weighted CE (ablation) | TMVA-Net · CARRADA-Test range-Doppler | mIoU (%): 51.1 -> 59.3 | Improved |
| add Soft Dice to weighted CE (ablation) | TMVA-Net · CARRADA-Test range-angle | mIoU (%): 34.3 -> 40.1 | Improved |
| add Soft Dice to plain CE (ablation) | TMVA-Net · CARRADA-Test range-Doppler | mIoU (%): 56.1 -> 45.2 | Hurt |
Closest papers by meaning
- Learning Structurally Consistent Representations for Multi-View Radar Semantic Segmentation (ECCV 2026 · similarity 0.88)
- Exploiting Temporal Relations on Radar Perception for Autonomous Driving (CVPR 2022 · similarity 0.88)
- RadarMOSEVE: A Spatial-Temporal Transformer Network for Radar-Only Moving Object Segmentation and Ego-Velocity Estimation (AAAI 2024 · similarity 0.86)
- V2X-Radar: A Multi-modal Dataset with 4D Radar for Cooperative Perception (NeurIPS 2025 · similarity 0.85)
- Bootstrapping Autonomous Driving Radars with Self-Supervised Learning (CVPR 2024 · similarity 0.84)
- TARSS-Net: Temporal-Aware Radar Semantic Segmentation Network (NeurIPS 2024 · 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.
