Real-Time Driver Drowsiness Detection for Embedded System Using Model Compression of Deep Neural Networks

Reddy, Bhargava; Kim, Ye-Hoon; Yun, Sojung; Seo, Chan-Won; Jang, Junik

doi:10.1109/cvprw.2017.59

Cited by 199 publications

(104 citation statements)

References 13 publications

Supporting

Mentioning

103

Contrasting

Unclassified

Order By: Relevance

“…All these troubling facts motivate the need for an economical solution that can detect drowsiness in early stages. It is commonly agreed [29,20,18] that there are three types of sources of information in drowsiness detection: Performance measurements, physiological measurements, and behavioral measurements.…”

Section: Introductionmentioning

confidence: 99%

“…Previous work in this field mostly focused on detecting extreme drowsiness with explicit signs such as yawning, nodding off and prolonged eye closure [19,20,25]. However, for drivers and workers, such explicit signs may not appear until only moments before an accident.…”

Section: Introductionmentioning

confidence: 99%

“…Previous work on drowsiness detection produced results on datasets that were either private [9] or acted [19,20]. By "acted" we mean data where subjects were instructed to simulate drowsiness, compared to "realistic" data, such as ours, where subjects were indeed drowsy in the corresponding videos.…”

Section: Introductionmentioning

confidence: 99%

“…By "acted" we mean data where subjects were instructed to simulate drowsiness, compared to "realistic" data, such as ours, where subjects were indeed drowsy in the corresponding videos. The lack of large, public, and realistic datasets has been pointed out by researchers in the field [18,19,20].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A Realistic Dataset and Baseline Temporal Model for Early Drowsiness Detection

Ghoddoosian

Galib

Athitsos

2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)

112

View full text Add to dashboard Cite

Drowsiness can put lives of many drivers and workers in danger. It is important to design practical and easy-todeploy real-world systems to detect the onset of drowsiness. In this paper, we address early drowsiness detection, which can provide early alerts and offer subjects ample time to react. We present a large and public real-life dataset 1 of 60 subjects, with video segments labeled as alert, low vigilant, or drowsy. This dataset consists of around 30 hours of video, with contents ranging from subtle signs of drowsiness to more obvious ones. We also benchmark a temporal model 2 for our dataset, which has low computational and storage demands. The core of our proposed method is a Hierarchical Multiscale Long Short-Term Memory (HM-LSTM) network, that is fed by detected blink features in sequence. Our experiments demonstrate the relationship between the sequential blink features and drowsiness. In the experimental results, our baseline method produces higher accuracy than human judgment.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Realistic Dataset and Baseline Temporal Model for Early Drowsiness Detection

Ghoddoosian

Galib

Athitsos

2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)

112

View full text Add to dashboard Cite

show abstract

“…Reddy et al [16] proposed a compressed deep neural network model that can be deployed on an embedded board. The authors note that for their focus, the NTHU-drowsy dataset had an unsuitable capture angle and inappropriate class labels.…”

Section: A Driver Drowsiness Detection Systemsmentioning

confidence: 99%

Detecting inter-sectional accuracy differences in driver drowsiness detection algorithms

Ngxande

Tapamo

Burke

2020

2020 International SAUPEC/RobMech/PRASA Conference

View full text Add to dashboard Cite

Convolutional Neural Networks (CNNs) have been used successfully across a broad range of areas including data mining, object detection, and in business. The dominance of CNNs follows a breakthrough by Alex Krizhevsky which showed improvements by dramatically reducing the error rate obtained in a general image classification task from 26.2% to 15.4%. In road safety, CNNs have been applied widely to the detection of traffic signs, obstacle detection, and lane departure checking. In addition, CNNs have been used in data mining systems that monitor driving patterns and recommend rest breaks when appropriate. This paper presents a driver drowsiness detection system and shows that there are potential social challenges regarding the application of these techniques, by highlighting problems in detecting dark-skinned drivers faces. This is a particularly important challenge in African contexts, where there are more dark-skinned drivers. Unfortunately, publicly available datasets are often captured in different cultural contexts, and therefore do not cover all ethnicities, which can lead to false detections or racially biased models. This work evaluates the performance obtained when training convolutional neural network models on commonly used driver drowsiness detection datasets and testing on datasets specifically chosen for broader representation. Results show that models trained using publicly available datasets suffer extensively from over-fitting, and can exhibit racial bias, as shown by testing on a more representative dataset. We propose a novel visualisation technique that can assist in identifying groups of people where there might be the potential of discrimination, using Principal Component Analysis (PCA) to produce a grid of faces sorted by similarity, and combining these with a model accuracy overlay.

show abstract