Research map / Video / Video segmentation and tracking
RaVF: Learning Radar Velocity Fields via Spatial-Doppler Guidance
SHENGPENG WANG, Lingzhen Li, Chunshen Li, Fei Xiao, Wei Wang
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes RaVF, a framework that learns scene-level velocity fields from 4D millimeter-wave radar using spatial-Doppler guidance to handle point sparsity/noise and Doppler ambiguity.
Problem
Scene-level velocity field estimation from 4D millimeter-wave radar is hard because point-based representations are sparse and noisy, and Doppler ambiguity from the Nyquist sampling limit breaks the continuity of velocity field learning.
Approach
Directly learns a velocity field from radar data guided jointly by spatial and Doppler cues, rather than relying on sparse point-based estimates, to address sampling sparsity and Doppler aliasing. (Details beyond the abstract not available.)
Closest papers by meaning
- RaUF: Learning the Spatial Uncertainty Field of Radar (CVPR 2026 · similarity 0.89)
- Full-Velocity Radar Returns by Radar-Camera Fusion (ICCV 2021 · similarity 0.86)
- Rethinking Radar 3D Perception: Representation Learning from Raw Spectra (NeurIPS 2026 · similarity 0.86)
- NEAR: Neural Electromagnetic Array Response (ICML 2025 · similarity 0.84)
- DART: Implicit Doppler Tomography for Radar Novel View Synthesis (CVPR 2024 · similarity 0.83)
- RadarMOSEVE: A Spatial-Temporal Transformer Network for Radar-Only Moving Object Segmentation and Ego-Velocity Estimation (AAAI 2024 · 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.
