Research map / Detection and segmentation / CNN features
RAVEN: Radar Adaptive Vision Encoders for Efficient Chirp-wise Object Detection and Segmentation
Anuvab Sen, Mir Sayeed Mohammad, Saibal Mukhopadhyay
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 |
|---|---|---|---|
| stop early using the calibrated chirp-state stopping rule (sub-frame instead of full frame) | RAVEN · RADIal | mIoU: 0.90 -> 0.85 | Hurt |
| stop early using the calibrated chirp-state stopping rule | RAVEN · RADIal | F1: 0.93 -> 0.89 | Hurt |
| stop early using the calibrated chirp-state stopping rule | RAVEN · RADIal | GMACs: 1.02 -> 0.27 | Improved |
| stop early using the calibrated chirp-state stopping rule | RAVEN · RADIal | latency (ms): 20.08 -> 9.15 | Improved |
| reduce chirp budget from 256 to 32-64 chirps | RAVEN · RADIal | speed/accuracy: 256 -> 32-64 chirps, more than 2x speedup | Improved |
| use chirp-wise SSM with cross-antenna attention instead of FFT-based frame backbone (vs TransRadar) | RAVEN (Full Frame) · RADIal | GMACs: 171.5 -> 1.02 | Improved |
| use RAVEN instead of T-FFTRadNet | RAVEN (Full Frame) · RADIal | mIoU: 0.79 -> 0.90 | Improved |
| use RAVEN instead of FFT-RadNet | RAVEN · RaDICaL | GMACs: 41.74 -> 0.053 | Improved |
| use RAVEN instead of FFT-RadNet | RAVEN · RaDICaL | Chamfer distance: 0.076 -> 0.082 | Hurt |
| use RAVEN instead of FFT-RadNet | RAVEN · RaDICaL | Dice: 0.996 -> 0.997 | No clear effect |
Closest papers by meaning
- AdaPKC: PeakConv with Adaptive Peak Receptive Field for Radar Semantic Segmentation (NeurIPS 2024 · similarity 0.82)
- RobuRCDet: Enhancing Robustness of Radar-Camera Fusion in Bird's Eye View for 3D Object Detection (ICLR 2025 · similarity 0.82)
- RETR: Multi-View Radar Detection Transformer for Indoor Perception (NeurIPS 2024 · similarity 0.82)
- CRKD: Enhanced Camera-Radar Object Detection with Cross-modality Knowledge Distillation (CVPR 2024 · similarity 0.82)
- RadarMOSEVE: A Spatial-Temporal Transformer Network for Radar-Only Moving Object Segmentation and Ego-Velocity Estimation (AAAI 2024 · similarity 0.82)
- SCKD: Semi-Supervised Cross-Modality Knowledge Distillation for 4D Radar Object Detection (AAAI 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.
