Synthetic FMCW Radar Range Azimuth Maps Augmentation with Generative Diffusion Model
cs.CV, cs.AI
Submitted: 2026-01-09
Updated: 2026-01-09
Journal ref: Proc. IEEE RadarConf, 2026
DOI: 10.1109/RadarConf2663773.2026.11675434
License: http://creativecommons.org/licenses/by-nc-sa/4.0/
The gist: The scarcity and low diversity of well-annotated automotive radar datasets often limit the performance of deep-learning-based environmental perception.
Terminology
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 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.
Sources
- Rethinking of Radar's Role: A Camera-Radar Dataset and Systematic Annotator via Coordinate Alignment
- K-Radar: 4D Radar Object Detection for Autonomous Driving in Various Weather Conditions
- Image-to-Image Translation with Conditional Adversarial Networks
- Denoising Diffusion Probabilistic Models
- Adversarial Supervision Makes Layout-to-Image Diffusion Models Thrive
- Semantic Image Synthesis via Diffusion Models
- L2RDaS: Synthesizing 4D Radar Tensors for Model Generalization via Dataset Expansion
- 4DR P2T: 4D Radar Tensor Synthesis with Point Clouds
- Auto-Encoding Variational Bayes
- Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
- Adding Conditional Control to Text-to-Image Diffusion Models
- Diffusion Model with Perceptual Loss
- Focal Loss for Dense Object Detection
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs