YOLOv12-Based Driver Monitoring System: Real-Time Applications in ADAS
Bibliographic record
Abstract
Driver Monitoring Systems (DMS) have become a cornerstone of Advanced Driver Assistance Systems (ADAS), aiming to reduce accidents caused by distraction, drowsiness, and unsafe driving behaviors.Traditional monitoring approaches, often reliant on indirect vehicle signals, have proven insufficient for real-time accuracy.With the advancement of deep learning and computer vision, vision-based systems now provide direct and reliable monitoring of driver states.This paper proposes a YOLOv12-based DMS designed for real-time, multi-class driver behavior recognition, addressing critical challenges such as low latency, false positives, and robustness under diverse environmental conditions.A custom dataset of 17,000 annotated images across four categories-Safe Driving, Distracted, Drowsy, and Smartphone Usage-was employed for training and evaluation.The proposed model achieved a precision of 95.0%, a recall of 92.5%, an F1-score of 93.7%, mAP@0.5 of 96.0%, and mAP@0.5:0.95 of 75.0%, outperforming prior YOLO versions under identical experimental conditions.These results demonstrate the superior accuracy, efficiency, and generalization capacity of YOLOv12 in handling complex visual cues, including partial occlusions and variable lighting.The system's strong performance underscores its potential as a reliable, real-time driver monitoring solution within ADAS, enabling proactive safety interventions and contributing significantly to the reduction of human-error-induced accidents on roads.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.011 | 0.003 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".