An Adjustable Gaze Tracking System and Its Application for Automatic Discrimination of Interest Objects

Adiba, Amalia Istiqlali; Tanaka, Nobuyuki; Miyake, Jun

doi:10.1109/tmech.2015.2470522

Cited by 7 publications

(2 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…O'Reilly 等 [12] 采用单个摄像机和多个光源, 通过半角估计和斜线距离 2 种方法求解眼球中心坐标, 用于屏幕 3D 视线估计. Liu 等 [13] 使用投影仪幕布进行平面注视点标定, 用于远距离的视线跟踪. Adiba 等 [14] 使用头戴式设备, 将注视点映射到场景摄像机的画面中, 并对场景中的物体识别分类. Kwon 等 [15] 通过左右眼图像及场景摄像机的对极几何原理, 用于驾驶环境中估计驾驶员视线在场景图中的视线落点. Sugano 等 [16] 通过大量带有脸部地标点标注的图像进行训练, 并通过随机森林算法对视线方向分类, 以实现 3D 视线方向估计.…”

Section: 相关工作unclassified

Long Distance 3D Gaze Estimation Combined with Radar Target Detection

Zhang¹,

Duan²,

Zhu³

et al. 2020

J Comput-Aided Design Comput Graphics

View full text Add to dashboard Cite

The 3D gaze estimation method based on 2D mapping and two lines of sight intersection has high accuracy in angle estimation. However, there are still large errors in the estimation of gaze distance. Thus, a 3D gaze estimation method based on an auxiliary millimeter wave radar target detection is proposed. First, it uses a movable calibration object to form a virtual calibration plane to calibrate the 2D planar fixation point.Next, the 2D fixation points of the right and left eyes on the calibration plane are calculated by the two-dimensional mapping function, and the 3D gaze direction is calculated combining the position of the eyes. Finally, the candidate point which has the smallest error from the gaze angle is selected from the target points detected by radar. A virtual plane is constructed at the distance of this point, by which the left and right lines of sight are intercept, the midpoint of the two intercept points is the estimation result of 3D fixation point. An experimental device is designed to fuse millimeter wave radar, camera and mobile calibration target. The 3D gaze point is tested in the range of 4.0~10.2 m. Experiments shows that with this method, the average angle error is 0.9°, the average Euclidean distance error is 130 mm. The result is better than binocular data geometric mapping method and depth camera aided estimation method.

show abstract

Section: 相关工作unclassified

Long Distance 3D Gaze Estimation Combined with Radar Target Detection

Zhang¹,

Duan²,

Zhu³

et al. 2020

J Comput-Aided Design Comput Graphics

View full text Add to dashboard Cite

show abstract

“…While research on wearable and embedded systems has increased significantly in recent years [17]- [19], only a small body of research has attempted to fuse muscle activity and motion data capture outside laboratory/clinical environments to exploit complementary aspects of each data set. These efforts, however, have focused principally on EMG, which narrows their collection time and range, leading to potentially unreliable signal quality over a prolonged time [20]- [22].…”

Section: Introductionmentioning

confidence: 99%

Pervasive Monitoring of Motion and Muscle Activation: Inertial and Mechanomyography Fusion

Woodward

Shefelbine

Vaidyanathan

2017

IEEE/ASME Trans. Mechatron.

View full text Add to dashboard Cite

Muscle activity and human motion are useful parameters to map the diagnosis, treatment, and rehabilitation of neurological and movement disorders. In laboratory and clinical environments, electromyography (EMG) and motion capture systems enable the collection of accurate, high resolution data on human movement and corresponding muscle activity. However, controlled surroundings limit both the length of time and the breadth of activities that can be measured. Features of movement, critical to understanding patient progress, can change during the course of a day and daily activities may not correlate to the limited motions examined in a laboratory. We introduce a system to measure motion and muscle activity simultaneously over the course of a day in an uncontrolled environment with minimal preparation time and ease of implementation that enables daily usage. Our system combines a bespoke inertial measurement unit (IMU) and mechanomyography (MMG) sensor, which measures the mechanical signal of muscular activity. The IMU can collect data continuously, and transmit wirelessly, for up to 10 hours. We describe the hardware design and validation and outline the data analysis (including data processing and activity classification algorithms) for the sensing system. Furthermore, we present two pilot studies to demonstrate utility of the system, including activity identification in six able-bodied subjects with an accuracy of 98%, and monitoring motion/muscle changes in a subject with cerebral palsy and of a single leg amputee over extended periods (∼5 hours). We believe these results provide a foundation for mapping human muscle activity and corresponding motion changes over time, providing a basis for a range of novel rehabilitation therapies.

show abstract