A lightweight deep learning model for real-time in-vehicle driver distraction detection with low-latency inference
- 1IRIS Team, MIASI Laboratory, Faculty of Applied Sciences, Nador, Mohammed Premier University
Abstract
Driver distraction is a major road-safety concern that requires reliable and efficient in-vehicle monitoring systems. The main contribution of this work is a reproducible driver-disjoint and deployment-oriented evaluation framework that jointly examines unseen-driver generalization, lightweight model benchmarking, explainability, calibration, and embedded inference. Experiments on the State Farm Distracted Driver Detection dataset show that MobileNetV3-Large provides the best trade-off among the evaluated lightweight models, achieving 88.92% test accuracy, 89.02% balanced accuracy, 88.07% macro-F1, and 97.88% Top-3 accuracy on unseen drivers. Explainable AI analysis indicates that the model mainly focuses on behavior-relevant regions, including the hands, face, phone area, steering wheel, and upper-body posture. For embedded deployment, TensorRT optimization on the Jetson Orin Nano Super achieved 212.94 FPS with 4.67 ms end-to-end latency in FP16 mode. These results demonstrate a practical balance between unseen-driver generalization, interpretability, and real-time embedded inference.
Related
- Deep learning-based detection of construction waste in complex scenarios with an improved lightweight algorithm
- Investigating the effects of acute caffeine administration on visual search, resistance to distraction, and attentional learning.
- Automatic segmentation of histological zones in rat distraction osteogenesis using a self-configuring deep learning pipeline
- TCLNet: A Thermal Contrast-Aware Lightweight Network for infrared object detection
- Subject-independent Parkinson's Disease Detection through A Lightweight Spectrotemporal Hybrid Network.
BibTeX
@article{essahraui2026lightweight,
title = {{A lightweight deep learning model for real-time in-vehicle driver distraction detection with low-latency inference}},
author = {Siham Essahraui and Chaymae Rami and Khalid El Makkaoui and Ibrahim Ouahbi},
journal = {Scientific Reports},
year = {2026},
month = jul,
doi = {10.1038/s41598-026-62654-5},
url = {https://doi.org/10.1038/s41598-026-62654-5},
note = {CC BY},
}
RIS
TY - JOUR
TI - A lightweight deep learning model for real-time in-vehicle driver distraction detection with low-latency inference
AU - Siham Essahraui
AU - Chaymae Rami
AU - Khalid El Makkaoui
AU - Ibrahim Ouahbi
PY - 2026
DA - 2026/07/01
JO - Scientific Reports
DO - 10.1038/s41598-026-62654-5
UR - https://doi.org/10.1038/s41598-026-62654-5
AB - Driver distraction is a major road-safety concern that requires reliable and efficient in-vehicle monitoring systems. The main contribution of this work is a reproducible driver-disjoint and deployment-oriented evaluation framework that jointly examines unseen-driver generalization, lightweight model benchmarking, explainability, calibration, and embedded inference. Experiments on the State Farm Distracted Driver Detection dataset show that MobileNetV3-Large provides the best trade-off among the evaluated lightweight models, achieving 88.92% test accuracy, 89.02% balanced accuracy, 88.07% macro-F1, and 97.88% Top-3 accuracy on unseen drivers. Explainable AI analysis indicates that the model mainly focuses on behavior-relevant regions, including the hands, face, phone area, steering wheel, and upper-body posture. For embedded deployment, TensorRT optimization on the Jetson Orin Nano Super achieved 212.94 FPS with 4.67 ms end-to-end latency in FP16 mode. These results demonstrate a practical balance between unseen-driver generalization, interpretability, and real-time embedded inference.
SN - 2045-2322
ER -