Synthetic FMCW Radar Range Azimuth Maps Augmentation with Generative Diffusion Model

TL;DR

This paper introduces a diffusion model-based framework for synthesizing realistic radar Range-Azimuth maps to augment datasets, improving perception accuracy and model generalization in automotive radar applications.

Contribution

It presents a novel conditional generative diffusion approach with geometry and temporal conditioning for realistic radar data synthesis, enhancing dataset diversity and model performance.

Findings

Signal reconstruction improves by 3.6 dB PSNR.

Data augmentation increases mean Average Precision by 4.15%.

Generated radar data is physically plausible and diverse.

Abstract

The scarcity and low diversity of well-annotated automotive radar datasets often limit the performance of deep-learning-based environmental perception. To overcome these challenges, we propose a conditional generative framework for synthesizing realistic Frequency-Modulated Continuous-Wave radar Range-Azimuth Maps. Our approach leverages a generative diffusion model to generate radar data for multiple object categories, including pedestrians, cars, and cyclists. Specifically, conditioning is achieved via Confidence Maps, where each channel represents a semantic class and encodes Gaussian-distributed annotations at target locations. To address radar-specific characteristics, we incorporate Geometry Aware Conditioning and Temporal Consistency Regularization into the generative process. Experiments on the ROD2021 dataset demonstrate that signal reconstruction quality improves by \SI{3.6}{dB} in Peak Signal-to-Noise Ratio over baseline methods, while training with a combination of real and synthetic datasets improves overall mean Average Precision by 4.15% compared with conventional image-processing-based augmentation. These results indicate that our generative framework not only produces physically plausible and diverse radar spectrum but also substantially improves model generalization in downstream tasks.