International Conferences
| Efficient On-Chip Implementation of 4D Radar-Based 3D Object Detection on Hailo-8L |
| Year |
2025 |
| Month |
|
| Journal |
ITS ASIA PACIFIC FORUM 2025 |
| Author |
Woong-Chan Byun, Dong-Hee Paek, Seung-Hyun Song, Seung-Hyun Kong* |
|
4-Dimensional (4D) Radar has attracted attention in autonomous driving due to their 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) architecture requires 5D inputs, the Hailo-8L only supports 4D tensors, posing a significant challenge. To overcome this limitation, we introduce a tensor transformation 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. |
https://arxiv.org/abs/2505.00757
28회 연결
