A survey of driver activity recognition from cameras installed in a car
DOI:
https://doi.org/10.17721/1812-5409.2020/1-2.15Abstract
The article discuses approaches to solving the problem of determining the activity of the driver from the cameras installed in the cargiven the actve development of intelligent driver asistance systems in recent years. The aricle provides an overview of the main problems that arise for the driver while driving Main advances in autonomous drving are presented and the classification of types of autonomous vehicles is provided . Next, the methods of solving the identified problems are described. The main part of the article focuses on solving the problem of determining the state of the driver during driving. Reasons for usage of computer vision and machine learning approaches for soving this task are described. The basic paradigms of the solution of his problem - classification of images, classification of a video stream, detection of the basic points of a body of the driver on the image from the camera installed inside a car are investigated. Main ideas of every method are described. The approaches are evaluated with identification of main advantages and drawbacks of the presented methods.
Key words: driver activity, driver assistance system, camera, neural network.
Pages of the article in the issue: 89 - 94
Language of the article: Ukrainian
References
Yan C. Vision based driver behaviour analysis [Electronic resource] / Chao Yan // University of Liverpool. – 2016. – Available from https://cgi.csc.liv.ac.uk/~frans/CurrentResearch/Thesi/yan_March2016.pdf.
Tamas V. Real-Time Distracted Drivers Detection Using Deep Learning [Electronic resource] / V. Tamas, V. Maties // IET Intelligent Transport Systems (IET INTELL TRANSP SY). – 2019. – Available from: https://www.researchgate.net/publication/326740203_Real-time_Detection_of_Distracted_Driving_based_on_Deep_Learning.
Gebert P. End-to-end Prediction of Driver Intention using 3D Convolutional Neural Networks [Electronic resource] / P. Gebert, A. Roitberg, M. Haurilet // Conference: 2019 IEEE Intelligent Vehicles Symposium (IV). – 2019. – Available from: https://www.researchgate.net/publication/335497206_End-to-end_Prediction_of_Driver_Intention_using_3D_Convolutional_Neural_Networks.
Eraqi H. Driver Distraction Identification with an Ensemble of Convolutional Neural Networks [Electronic resource] / Hesham Eraqi // Journal of Advanced Transportation. – 2019. – Available from: https://arxiv.org/abs/1901.09097.
Cao Z. Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields [Electronic resource] / Z. Cao, T. Simon // 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2017. Available from: https://arxiv.org/abs/1611.08050.
Alioua N. Driver head pose estimation using efficient descriptor fusion [Electronic resource] / N. Alioua, A. Amine, A. Rogozan // EURASIP Journal on Image and Video Processing. – 2016. – Available from: https://jivp-eurasipjournals.springeropen.com/articles/10.1186/s13640-016-0103-z.
OpenPose: Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields [Electronic resource] / Z.Cao, G. Hidalgo Martinez, T. Simon, S. Wei // EEE Transactions on Pattern Analysis and Machine Intelligence. – 2019. – Available from: https://arxiv.org/abs/1812.08008.
Simonyan K. Very Deep Convolutional Networks for Large-Scale Image Recognition [Electronic resource] / K. Simonyan, A. Zisserman // ICLR 2015. – 2014. – Available from: https://arxiv.org/abs/1409.1556.
Ruiz N. Fine-Grained Head Pose Estimation Without Keypoints [Electronic resource] / N. Ruiz, E. Chong, J. M. Rehg // The IEEE Conference on Computer Vision and Pattern Recognition (CVPR). – 2017. – Available from: https://arxiv.org/abs/1710.00925.
Downloads
How to Cite
Issue
Section
License
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).