首次在Hailo-8L上实现4D雷达3D目标检测，低功耗实时运行。
Efficient On-Chip Implementation of 4D Radar-Based 3D Object Detection on Hailo-8L
- 通过张量转换将5D输入转为4D，适配Hailo-8L硬件限制。
- 达到46.47% AP_3D和52.75% AP_BEV，推理速度13.76 Hz。
- 适合自动驾驶中低功耗嵌入式雷达感知系统部署。
4D雷达因能在恶劣天气下实现鲁棒的3D目标检测而受到自动驾驶领域关注。为实现技术落地，需在低功耗嵌入式环境中实现实时处理。本文首次在Hailo-8L AI加速器上实现了4D雷达3D目标检测模型的片上部署。尽管传统3D卷积神经网络（CNN）需要5D输入，但Hailo-8L仅支持4D张量，构成重大挑战。为此，我们提出一种编译期张量转换方法，将5D输入重塑为4D格式，无需修改模型结构即可直接部署。所提系统在保持与GPU模型相当精度的同时，实现13.76 Hz的推理速度，获得46.47% AP_3D和52.75% AP_BEV。结果证明4D雷达感知技术可有效应用于自动驾驶系统。
原文摘要 · Abstract (English)
4D radar has attracted attention in autonomous driving due to its ability to enable robust 3D object detection even under adverse weather conditions. To practically deploy such technologies, it is essential to achieve real-time processing within low-power embedded environments. Addressing this, we present the first on-chip implementation of a 4D radar-based 3D object detection model on the Hailo-8L AI accelerator. Although conventional 3D convolutional neural network (CNN) architectures require 5D inputs, the Hailo-8L only supports 4D tensors, posing a significant challenge. To overcome this limitation, we introduce a tensor transformation method that reshapes 5D inputs into 4D formats during the compilation process, enabling direct deployment without altering the model structure. The proposed system achieves 46.47% AP_3D and 52.75% AP_BEV, maintaining comparable accuracy to GPU-based models while achieving an inference speed of 13.76 Hz. These results demonstrate the applicability of 4D radar-based perception technologies to autonomous driving systems.
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。