arXiv cs.AI· Olivia Nocentini, Marta Lagomarsino, Gokhan Solak, Younggeol Cho, Qiyi Tong, Sara Zeynalpour, Marta Lorenzini, Alessandro Ledda, Arash Ajoudani·· 5 小时前AI 评分36
基于人脸与上半身关键点的图神经网络识别模拟火车司机状态研究
Graph-Based Recognition of Simulated Train-Driver States From Facial and Upper-Body Keypoints
AI 导读
一项研究提出仅用单目前置 RGB 摄像头和图神经网络对模拟火车司机状态进行三分类：清醒、不清醒和扮演的紧急行为。消融实验显示，光照条件下融合人脸与骨骼特征的三分类准确率达 81%，清醒/不清醒二分类达 99%，均优于仅用单一特征。研究团队还发布了包含三种光照条件的受控 RGB 视频数据集。
正文 · 原文
Abstract:Driver fatigue poses a significant challenge to railway safety, with traditional systems like the dead-man switch offering limited and basic alertness checks. This study presents a vision-based monitoring system that relies solely on a single front-facing RGB camera and a graph neural network to classify simulated train-driver states into alert, not-alert, and an emergency class comprising acted emergency-like behaviours. To optimize input representations for the model, an ablation study was performed, comparing three feature configurations: skeletal-only, facial-only, and a combination of both. Experimental results show that combining facial and skeletal features yields the highest accuracy (81%) for the three-class model under the light condition, outperforming models that use only facial or skeletal features. Furthermore, the combination of facial and skeletal features achieves 99% accuracy in the alert/not alert classification in light condition. Additionally, we introduced a controlled RGB video dataset containing alert, not alert, and acted emergency-like behaviours recorded under three illumination conditions. These contributions represent a step toward passive and non-contact train-driver state recognition based on facial and upper-body dynamics.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.07083 [cs.CV] |
| (or arXiv:2610.07083v2 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2610.07083 arXiv-issued DOI via DataCite |
Submission history
From: Olivia Nocentini [view email]
[v1]
Mon, 5 Oct 2026 11:25:16 UTC (8,550 KB)
[v2]
Wed, 28 Oct 2026 19:05:05 UTC (7,037 KB)
来源：arXiv cs.AI · arxiv.org