Real-time Distracted Driver Posture Classification

Abouelnaga, Yehya; Eraqi, Hesham M.; Moustafa, Mohamed

doi:10.48550/arxiv.1706.09498

Cited by 25 publications

(52 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Driver's State/Gesture Recognition: The performance of RAN using eight different base networks and their comparison to the state-of-the-art approaches is presented in Table 4. For the dataset V1 [32], RAN with Inception-V3 [48] as a base CNN is the best (99.47%) performer consistent with that for head pose recognition in the last section. Moreover, the proposed RAN with most base networks (except VGG16) has outperformed the approach in [31], which is the best among existing works.…”

Section: Resultssupporting

confidence: 71%

“…For the dataset V1, the proposed RAN is compared with the VGG with regularization (R-VGG) and modified VGG (M-VGG) proposed in [31]. We also compare the performance with that of the Genetic Algorithm Weighted Ensemble (GA-WE) [32] in Fig 4a . It can be seen that in all the categories, our approach is better than the state-of-theart ones. Similarly, in dataset V2, the proposed RAN with DenseNet-169 as a base CNN is compared with the only available method, Inception-V3 [33], in class-wise accuracy.…”

Section: Resultsmentioning

confidence: 99%

“…Recently, there has been significant progress in monitoring driver's gesture/state using monocular images [28] [31] [32] [33]. Most of these approaches focus on human-centric cues such as body pose [28], body-object interactions [29], and hand positions and movements [32] [33]. Behera et al [28] propose a method for drivers state/gesture recognition by injecting latent body pose into the adapted DenseNet architecture [34].…”

Section: Driver's Gestures Recognitionmentioning

confidence: 99%

“…Baheti et al [31] modify the VGG16 architecture [35] to improve the driver's state classification accuracy by reducing the number of parameters for faster execution. Abouelnaga et al [32] achieve a high classification accuracy of driver's state/gesture by considering a genetically weighted ensemble of five different CNNs, making it too heavy for real-time applications, which are very much essential in autonomous/self-driving cars.…”

Section: Driver's Gestures Recognitionmentioning

confidence: 99%

“…We validate our model using two challenging datasets: i) Distracted Driver V1 [32], and ii) Distracted Driver V2 [33]. These datasets consist of 10 classes of actions: 1) safe driving, 2) texting -right, 3) talking on the phone -right, In our experiment, we use the VIS camera (10,379 images) with a strong illumination condition.…”

Section: Driver's State/gesture Recognition Datasetsmentioning

confidence: 99%

See 4 more Smart Citations

Regional Attention Network (RAN) for Head Pose and Fine-Grained Gesture Recognition

Behera¹,

Wharton²,

Ghahremani³

et al. 2023

IEEE Trans. Affective Comput.

View full text Add to dashboard Cite

Affect is often expressed via non-verbal body language such as actions/gestures, which are vital indicators for human behaviors. Recent studies on recognition of fine-grained actions/gestures in monocular images have mainly focused on modeling spatial configuration of body parts representing body pose, human-objects interactions and variations in local appearance. The results show that this is a brittle approach since it relies on the accurate body parts/objects detection. In this work, we argue that there exist local discriminative semantic regions, whose "informativeness" can be evaluated by the attention mechanism for inferring fine-grained gestures/actions. To this end, we propose a novel end-to-end Regional Attention Network (RAN), which is a fully Convolutional Neural Network (CNN) to combine multiple contextual regions through attention mechanism, focusing on parts of the images that are most relevant to a given task. Our regions consist of one or more consecutive cells and are adapted from the strategies used in computing HOG (Histogram of Oriented Gradient) descriptor. The model is extensively evaluated on ten datasets belonging to 3 different scenarios: 1) head pose recognition, 2) drivers state recognition, and 3) human action and facial expression recognition. The proposed approach outperforms the state-of-the-art by a considerable margin in different metrics.

show abstract

Section: Resultssupporting

confidence: 71%

Section: Resultsmentioning

confidence: 99%

Section: Driver's Gestures Recognitionmentioning

confidence: 99%

Section: Driver's Gestures Recognitionmentioning

confidence: 99%

Section: Driver's State/gesture Recognition Datasetsmentioning

confidence: 99%

See 3 more Smart Citations

Regional Attention Network (RAN) for Head Pose and Fine-Grained Gesture Recognition

Behera¹,

Wharton²,

Ghahremani³

et al. 2023

IEEE Trans. Affective Comput.

View full text Add to dashboard Cite

show abstract

DriSm_YNet: a breakthrough in real-time recognition of driver smoking behavior using YOLO-NAS

Pandey,

Pati,

Maurya

2024

Neural Comput & Applic

View full text Add to dashboard Cite

Driver Distracted Behavior Detection Using a Light Weight Model based on the W-MSA

Lyu

2023

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

The percentage of traffic accidents caused by driver factors is about 90% in the world. Despite the great development of autonomous driving, it is still not completely self-driving. So, it is still not possible to avoid traffic accidents caused by drivers. Computer vision technology has made great progress with deep learning development. That makes it possible to detect the driver’s behaviour using a camera. To reduce the detection price, this paper presents a light weight model to detect the driver’s behavior based on the W-MSA. This model consists of 2 encoder modules and a classification module. And it used the Global Avgpool and W-MSA to reduce the model parameter and FLOPs. To avoid the low accuracy of the detection, this paper also used label smoothing regularization and CBAM technologies to improve the accuracy. This paper also used a visualization method to show the interpretability of the proposed model. The results show that the accuracy of the proposed model is 98% on the Kaggle driving test dataset. Compared to other state-of-the-art models, our method has a high accuracy with fewer model parameters.

show abstract

Real-time Distracted Driver Posture Classification

Cited by 25 publications

References 0 publications

Regional Attention Network (RAN) for Head Pose and Fine-Grained Gesture Recognition

Regional Attention Network (RAN) for Head Pose and Fine-Grained Gesture Recognition

DriSm_YNet: a breakthrough in real-time recognition of driver smoking behavior using YOLO-NAS

Driver Distracted Behavior Detection Using a Light Weight Model based on the W-MSA

Contact Info

Product

Resources

About