用扩散模型生成高质量4D雷达点云，解决标注数据少的问题。
4D-RaDiff: Latent Diffusion for 4D Radar Point Cloud Generation
- 在隐空间中对稀疏雷达点云进行扩散生成，支持物体与场景级控制。
- 合成数据可使检测性能提升，且减少90%真实标注数据需求。
- 适合自动驾驶感知系统训练，尤其适用于雷达数据稀缺场景。
车载雷达因其成本低、恶劣天气下表现稳定，在环境感知中展现出广阔前景。然而，标注过的雷达数据极为有限，制约了雷达感知系统的发展。为此，我们提出一种新框架——4D-RaDiff，用于生成4D雷达点云以用于目标检测器的训练与评估。不同于图像扩散模型，该方法针对雷达点云的稀疏性与独特特征，将扩散过程应用于隐空间点云表示，并支持在物体或场景层面进行条件控制。4D-RaDiff能将未标注的边界框转化为高质量雷达标注，并将现有激光雷达点云转换为逼真的雷达场景。实验表明，将4D-RaDiff生成的合成数据作为数据增强手段，可显著提升检测性能；此外，基于合成数据预训练可将所需标注雷达数据量减少高达90%，同时保持相当的检测性能。
原文摘要 · Abstract (English)
Automotive radar has shown promising developments in environment perception due to its cost-effectiveness and robustness in adverse weather conditions. However, the limited availability of annotated radar data poses a significant challenge for advancing radar-based perception systems. To address this limitation, we propose a novel framework to generate 4D radar point clouds for training and evaluating object detectors. Unlike image-based diffusion, our method is designed to consider the sparsity and unique characteristics of radar point clouds by applying diffusion to a latent point cloud representation. Within this latent space, generation is controlled via conditioning at either the object or scene level. The proposed 4D-RaDiff converts unlabeled bounding boxes into high-quality radar annotations and transforms existing LiDAR point cloud data into realistic radar scenes. Experiments demonstrate that incorporating synthetic radar data of 4D-RaDiff as data augmentation method during training consistently improves object detection performance compared to training on real data only. In addition, pre-training on our synthetic data reduces the amount of required annotated radar data by up to 90% while achieving comparable object detection performance.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。