Fig 1 - available via license: Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International
Content may be subject to copyright.
Class distribution of the complete DMD dataset. Both geometrical and temporal-based annotation are included in the available DMD ground truth data.
Source publication
Vision is the richest and most cost-effective technology for Driver Monitoring Systems (DMS), especially after the recent success of Deep Learning (DL) methods. The lack of sufficiently large and comprehensive datasets is currently a bottleneck for the progress of DMS development, crucial for the transition of automated driving from SAE Level-2 to...
Contexts in source publication
Context 1
... for gaze allocation to interior regions, head-pose estimation and different driver's hands' positions and interaction with inside objects (i.e. steering wheel, bottle, cellphone, etc.). There was a participation of 37 volunteers for this experiment, the gender proportions are 73% and 27% for men and women, respectively, 10 wearing glasses (see Fig. 1). The age distribution of the participants was homogeneous in the range of 18 to 50 years. The participants were selected to assure novice and expert drivers were included in the recordings. Each participant signed an GDPR informed consent which allows the dataset to be publicly available for research purposes. Moreover, for certain ...
Context 2
... was done according to the recording type and could include: geometrical features (i.e. landmarks points and bounding boxes), temporal features (i.e. events and actions) or context. Each driver behaviour type has a diverse set of both geometric and temporal classes. Fig. 1 depicts the full distribution of classes available in the DMD. This distribution for the different types of recordings was done using a total of 93 classes (temporal, geometric and context). In this paper we focus on a subset of the distraction scenario in which we annotate temporal actions used by the DL algorithm to identify driver ...
Citations
... This model has the potential to be employed in real-time distracted driver detection systems. In 2020, a new Driver Monitoring Dataset (DMD) was introduced [6], and the authors of [7] utilized this dataset for their research. DMD consisted of recordings taken from three inside-the-car viewpoints, each having three cameras positioned to take pictures of the driver's hands, face, and body. ...
Researchers are concentrating on developing technologies to identify and caution drivers against driving while distracted because it is a major cause of traffic accidents. According to the National Highway Traffic Safety Administrator's report, distracted driving is to blame for roughly one in every five car accidents.Our goal is to create an accurate and dependable method for identifying distracted drivers and alerting them to their lack of focus. We take inspiration from the success of convolutional neural networks in computer vision to do this. Our strategy entails putting in place a CNN-based system that can recognize when a driver is distracted as well as pinpoint the precise cause of their preoccupation. Real-time detection, however, necessitates three apparently mutually exclusive requirements for an optimal network: a small number of parameters, high accuracy, and fast speed.
