Driver Gaze Region Estimation Based on Computer Vision

Shan, Xinghua; Wang, Zengcai; Liu, Xiangpeng; Lin, Mingxing; Zhao, Lei; Wang, Jinshan; Wang, Guodong

doi:10.1109/icmtma50254.2020.00085

Cited by 5 publications

(6 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…2, many researchers have attempted to develop contactless monitoring systems for various applications. Previous methods for monitoring people can be divided into 1) vision-based [4][5][6][7][8][9][10][11][12][13], 2) wearable-based [14][15][16][17][18][19][20][21], and 3) head-gaze-based technologies [22][23][24][25][26][27][28][29][30][31]. Vision-based methods usually detect and track the pose or movements of medically vulnerable people using user images captured by cameras.…”

Section: Figure 1: Examples Of Medically Vulnerable Peoplementioning

confidence: 99%

“…For example, head-gaze estimation systems estimate where a user is looking (e.g., a side or rear mirror, etc.) while driving a vehicle [22][23][24][25][26][27][28][29][30][31] (refer to Fig. 3b).…”

Section: Figure 1: Examples Of Medically Vulnerable Peoplementioning

confidence: 99%

“…As stated in the Introduction, head-gaze is a proxy of user gaze direction computed based on the head orientation of users, a useful indicator that does not require precise calibration to represent the interests and intentions of users [22]. Previous studies on the estimation of headgaze mainly focused on monitoring the status of users (e.g., whether the user is concentrating) while driving [22][23][24][25][26][27][28][29][30][31]. These studies first divided the in-vehicle space into multiple zones (e.g., right mirror, rear-view mirror, and speed meter areas).…”

Section: Head-gaze Estimationmentioning

confidence: 99%

“…Afterward, a single camera was installed on the dashboard of a vehicle to capture RGB images of the driver's face while looking at certain in-vehicle zones. Finally, the authors of [23][24][25][26][27] trained a deep learning model and conventional machine learning algorithms using the captured RGB images to estimate the head-gaze direction. In addition, the authors of [28][29][30] collected the depth, infrared images, or 3D point cloud data of a driver, which are not primarily affected by lighting conditions, to estimate the direction of the driver's head-gaze.…”

Section: Head-gaze Estimationmentioning

confidence: 99%

See 3 more Smart Citations

Multi-View Multi-Modal Head-Gaze Estimation for Advanced Indoor User Interaction

Kim¹,

Jeong²

2022

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

Gaze estimation is one of the most promising technologies for supporting indoor monitoring and interaction systems. However, previous gaze estimation techniques generally work only in a controlled laboratory environment because they require a number of high-resolution eye images. This makes them unsuitable for welfare and healthcare facilities with the following challenging characteristics: 1) users' continuous movements, 2) various lighting conditions, and 3) a limited amount of available data. To address these issues, we introduce a multi-view multi-modal head-gaze estimation system that translates the user's head orientation into the gaze direction. The proposed system captures the user using multiple cameras with depth and infrared modalities to train more robust gaze estimators under the aforementioned conditions. To this end, we implemented a deep learning pipeline that can handle different types and combinations of data. The proposed system was evaluated using the data collected from 10 volunteer participants to analyze how the use of single/multiple cameras and modalities affect the performance of head-gaze estimators. Through various experiments, we found that 1) an infrared-modality provides more useful features than a depth-modality, 2) multi-view multi-modal approaches provide better accuracy than singleview single-modal approaches, and 3) the proposed estimators achieve a high inference efficiency that can be used in real-time applications.

show abstract

Section: Figure 1: Examples Of Medically Vulnerable Peoplementioning

confidence: 99%

Section: Figure 1: Examples Of Medically Vulnerable Peoplementioning

confidence: 99%

Section: Head-gaze Estimationmentioning

confidence: 99%

Section: Head-gaze Estimationmentioning

confidence: 99%

See 2 more Smart Citations

Multi-View Multi-Modal Head-Gaze Estimation for Advanced Indoor User Interaction

Kim¹,

Jeong²

2022

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

show abstract

“…Ref. [ 22 ] based their algorithm on detecting head pose and landmark detection. They based the head pose algorithm on Posing from Orthography and Scaling with Iterations (POSIT), which assesses the 3D posture of an object in a camera frame.…”

Section: Related Workmentioning

confidence: 99%

Single Camera Face Position-Invariant Driver’s Gaze Zone Classifier Based on Frame-Sequence Recognition Using 3D Convolutional Neural Networks

Lollett

Kamezaki

Sugano

2022

Sensors

View full text Add to dashboard Cite

Estimating the driver’s gaze in a natural real-world setting can be problematic for different challenging scenario conditions. For example, faces will undergo facial occlusions, illumination, or various face positions while driving. In this effort, we aim to reduce misclassifications in driving situations when the driver has different face distances regarding the camera. Three-dimensional Convolutional Neural Networks (CNN) models can make a spatio-temporal driver’s representation that extracts features encoded in multiple adjacent frames that can describe motions. This characteristic may help ease the deficiencies of a per-frame recognition system due to the lack of context information. For example, the front, navigator, right window, left window, back mirror, and speed meter are part of the known common areas to be checked by drivers. Based on this, we implement and evaluate a model that is able to detect the head direction toward these regions having various distances from the camera. In our evaluation, the 2D CNN model had a mean average recall of 74.96% across the three models, whereas the 3D CNN model had a mean average recall of 87.02%. This result show that our proposed 3D CNN-based approach outperforms a 2D CNN per-frame recognition approach in driving situations when the driver’s face has different distances from the camera.

show abstract

Towards a Driver's Gaze Zone Classifier using a Single Camera Robust to Temporal and Permanent Face Occlusions

Lollett

Kamezaki

Sugano

2021

2021 IEEE Intelligent Vehicles Symposium (IV)

View full text Add to dashboard Cite

Driver Gaze Region Estimation Based on Computer Vision

Cited by 5 publications

References 11 publications

Multi-View Multi-Modal Head-Gaze Estimation for Advanced Indoor User Interaction

Multi-View Multi-Modal Head-Gaze Estimation for Advanced Indoor User Interaction

Single Camera Face Position-Invariant Driver’s Gaze Zone Classifier Based on Frame-Sequence Recognition Using 3D Convolutional Neural Networks

Towards a Driver's Gaze Zone Classifier using a Single Camera Robust to Temporal and Permanent Face Occlusions

Contact Info

Product

Resources

About