Baseline face detection, head pose estimation, and coarse direction detection for facial data in the SHRP2 naturalistic driving study

Paone, Jeffrey R.; Bolme, David S.; Ferrell, Regina; Aykac, Deniz; Karnowski, Thomas P.

doi:10.1109/ivs.2015.7225682

Cited by 21 publications

(15 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Acquisition scenario covers the circumstances under which the acquisition of the head pose takes place. The most important distinction is between in-laboratory [4,7,8,11]- [13] vs. in-the-wild [5,6,9,10,14] acquisition. While the former restricts the data by defining a rather well-defined, static environment, the latter offers more variety through being acquired in unconstrained environments such as outside, thus covering many challenges like differing illumination and variable background.…”

Section: Related Workmentioning

confidence: 99%

“…By using an orthographic projection, this approach only allows for approximate position and orientation estimates. NDS-HPV [9] contains 2PB of highly compressed, low resolution images from a naturalistic driving study. It contains images of over 3100 subjects collected over a period of over 2 years.…”

Section: B Driving Head Pose Datasetsmentioning

confidence: 99%

“…(a) Bosphorus [4] (b) Biwi Kinect [8] (c) ICT-3DHP [7] (d) gi4e hpdb [11] (e) SynHead [12] (f) UbiPose [13] (g) RS-DMV [5] (h) Lisa-P [6] (i) NDS HPV [9] (j) VIVA [10] (k) DriveAHead [14]. Content within red box provided.…”

Section: Dd-pose -A Large-scale Driver Head Pose Benchmarkmentioning

confidence: 99%

See 2 more Smart Citations

DD-Pose - A large-scale Driver Head Pose Benchmark

Roth

Gavrila

2019

2019 IEEE Intelligent Vehicles Symposium (IV)

View full text Add to dashboard Cite

We introduce DD-Pose, the Daimler TU Delft Driver Head Pose Benchmark, a large-scale and diverse benchmark for image-based head pose estimation and driver analysis. It contains 330k measurements from multiple cameras acquired by an in-car setup during naturalistic drives. Large out-of-plane head rotations and occlusions are induced by complex driving scenarios, such as parking and driver-pedestrian interactions. Precise head pose annotations are obtained by a motion capture sensor and a novel calibration device. A high resolution stereo driver camera is supplemented by a camera capturing the driver cabin. Together with steering wheel and vehicle motion information, DD-Pose paves the way for holistic driver analysis. Our experiments show that the new dataset offers a broad distribution of head poses, comprising an order of magnitude more samples of rare poses than a comparable dataset. By an analysis of a state-of-the-art head pose estimation method, we demonstrate the challenges offered by the benchmark. The dataset and evaluation code are made freely available to academic and non-profit institutions for non-commercial benchmarking purposes.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: B Driving Head Pose Datasetsmentioning

confidence: 99%

See 1 more Smart Citation

DD-Pose - A large-scale Driver Head Pose Benchmark

Roth

Gavrila

2019

2019 IEEE Intelligent Vehicles Symposium (IV)

View full text Add to dashboard Cite

show abstract

“…The eye tracker has two cameras: one records the eye movement (monocular), and the other records the front view of the subject. The eye tracker is able to measure the gaze point which means it may detect the time when the subject take his/her vision away from the road, and the time of eye closure which is a common method in drowsiness detection [32][33] [34]. However, the eye tracker is relatively expensive and requires training for the operator to be able to use it properly.…”

Section: B Pilot Experiments IImentioning

confidence: 99%

“…The failure to employ an eye tracker in this pilot study made such analysis at the moment is impossible. With regard to eye movements recording, it is possible to use a video camera as replacement of eye tracker [32]. The video camera is cheap and easy to use.…”

Section: E Procedures For Further Experimentsmentioning

confidence: 99%

Preliminary investigation of sleep-related driving fatigue experiment in Indonesia

Sanjaya

Sya'Bana

Hutchinson

et al. 2018

J. Mechatron. Electr. Power Veh. Technol.

View full text Add to dashboard Cite

Sleep-related driving fatigue has been recognised as one main cause of traffic accidents. In Indonesia, experiment-based driving fatigue study is still very limited, therefore it is necessary to develop laboratory-based experiment procedure for sleep-related fatigue study. In this preliminary study, we performed a literature review to find references for the procedure and three pilot experiments to test the instruments and procedure to be used in measuring driving fatigue. Three subjects participated, both from experienced and inexperienced drivers. Our pilot experiments were performed on a driving simulator using OpenDS software with brake and lane change test reaction time measurement. We measured sleepiness by using Karolinska Sleepiness Scale (KSS) Questionnaire. The conditions of the experiment were based on illumination intensity as well as pre- and post-lunch session. We found that lane change reaction time is more potential than brake reaction time to measure driving performance as shown by the more fluctuating data. Post-lunch seems to induce drowsiness greater than illumination intensity. KSS questionnaire seems non-linear with driving performance data. We need to test further these speculations in the future studies involving a sufficient number of subjects. We also need to compare the effect of circadian rhythm and sleep deprivation on driving fatigue. The use of eye closure and physiological measurement in further study will enable us to measure driving fatigue more objectively. Considering the limitations, more preliminary experiments are required to be performed before conducting the main experiment of driving fatigue.

show abstract

Support Vector Machine-Based Face Direction Detection Using an Infrared Array Sensor

Chen

Liu

2018

Conference Proceedings of the Society for Experimental Mechanics Series

View full text Add to dashboard Cite

Baseline face detection, head pose estimation, and coarse direction detection for facial data in the SHRP2 naturalistic driving study

Cited by 21 publications

References 17 publications

DD-Pose - A large-scale Driver Head Pose Benchmark

DD-Pose - A large-scale Driver Head Pose Benchmark

Preliminary investigation of sleep-related driving fatigue experiment in Indonesia

Support Vector Machine-Based Face Direction Detection Using an Infrared Array Sensor

Contact Info

Product

Resources

About