RadarGen: Automotive Radar Point Cloud Generation from Cameras
This work addresses the need for scalable multimodal generative simulation in autonomous driving, though it is incremental by adapting existing diffusion methods to the radar domain.
The paper tackles the problem of generating realistic automotive radar point clouds from multi-view camera images using a diffusion model, resulting in improved alignment with real data distributions and reduced gaps for perception models.
We present RadarGen, a diffusion model for synthesizing realistic automotive radar point clouds from multi-view camera imagery. RadarGen adapts efficient image-latent diffusion to the radar domain by representing radar measurements in bird's-eye-view form that encodes spatial structure together with radar cross section (RCS) and Doppler attributes. A lightweight recovery step reconstructs point clouds from the generated maps. To better align generation with the visual scene, RadarGen incorporates BEV-aligned depth, semantic, and motion cues extracted from pretrained foundation models, which guide the stochastic generation process toward physically plausible radar patterns. Conditioning on images makes the approach broadly compatible, in principle, with existing visual datasets and simulation frameworks, offering a scalable direction for multimodal generative simulation. Evaluations on large-scale driving data show that RadarGen captures characteristic radar measurement distributions and reduces the gap to perception models trained on real data, marking a step toward unified generative simulation across sensing modalities.