Radar Fields: Frequency-Space Neural Scene Representations for FMCW Radar
This work addresses the lack of neural reconstruction methods for radar, a cost-effective and weather-robust sensing modality important for autonomous vehicles and robots operating outdoors, representing a novel application rather than an incremental improvement.
The paper tackles the problem of neural scene reconstruction for radar data, which is robust in adverse weather, by introducing Radar Fields, a method that learns fields in Fourier frequency space to synthesize raw radar measurements and extract scene occupancy, achieving effective results across diverse outdoor scenarios including harsh weather conditions.
Neural fields have been broadly investigated as scene representations for the reproduction and novel generation of diverse outdoor scenes, including those autonomous vehicles and robots must handle. While successful approaches for RGB and LiDAR data exist, neural reconstruction methods for radar as a sensing modality have been largely unexplored. Operating at millimeter wavelengths, radar sensors are robust to scattering in fog and rain, and, as such, offer a complementary modality to active and passive optical sensing techniques. Moreover, existing radar sensors are highly cost-effective and deployed broadly in robots and vehicles that operate outdoors. We introduce Radar Fields - a neural scene reconstruction method designed for active radar imagers. Our approach unites an explicit, physics-informed sensor model with an implicit neural geometry and reflectance model to directly synthesize raw radar measurements and extract scene occupancy. The proposed method does not rely on volume rendering. Instead, we learn fields in Fourier frequency space, supervised with raw radar data. We validate the effectiveness of the method across diverse outdoor scenarios, including urban scenes with dense vehicles and infrastructure, and in harsh weather scenarios, where mm-wavelength sensing is especially favorable.