{"slug": "graph-based-recognition-of-simulated-train-driver-states-from-facial-and-upper", "title": "Graph-Based Recognition of Simulated Train-Driver States From Facial and Upper-Body Keypoints", "summary": "A graph neural network combining facial and upper-body skeletal keypoints classified simulated train-driver states into alert, not-alert and emergency classes with 81% accuracy in light conditions, according to a study published on arXiv as 2610.07083v1. The same combined feature configuration reached 99% accuracy on the binary alert/not-alert task under light illumination, outperforming skeletal-only and facial-only models. The authors also released a controlled RGB video dataset of alert, not-alert and acted emergency-like behaviours recorded under three illumination conditions, aimed at passive, non-contact train-driver monitoring.", "body_md": "arXiv:2610.07083v1 Announce Type: new \nAbstract: 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.", "url": "https://wpnews.pro/news/graph-based-recognition-of-simulated-train-driver-states-from-facial-and-upper", "canonical_source": "https://arxiv.org/abs/2610.07083", "published_at": "2026-10-07 04:00:00+00:00", "updated_at": "2026-10-07 04:16:27.241438+00:00", "lang": "en", "topics": ["computer-vision", "machine-learning", "neural-networks", "ai-research"], "entities": ["arXiv", "graph neural network"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/graph-based-recognition-of-simulated-train-driver-states-from-facial-and-upper", "markdown": "https://wpnews.pro/news/graph-based-recognition-of-simulated-train-driver-states-from-facial-and-upper.md", "text": "https://wpnews.pro/news/graph-based-recognition-of-simulated-train-driver-states-from-facial-and-upper.txt", "jsonld": "https://wpnews.pro/news/graph-based-recognition-of-simulated-train-driver-states-from-facial-and-upper.jsonld"}}