Driver Hand Detection Using Squeeze-and-Excitation YOLOv4 Network
Zhe Ma, Xiaohui Yang, Yingji Zhang · 2020
In this paper, we have proposed a driver hand detection method using YOLOv4 combined with Squeeze-and-Excitation (SE) module, which we called SE-YOLOv4. In our proposed SE-YOLOv4, we add the SE module to the CSPDarknet in the backbone of YOLOv4. The SE module optimizes the learning ability of YOLOv4, which improves the performance of driver hand detection. Furthermore, we have established a driver hand detection dataset. We evaluate the proposed hand detection method and compare it with some other network model, and experimental results show that the hand detection accuracy of our method was improved by 2%-4%.