Acquisition of 3D gaze information from eyeball movements using inside-out camera

Shimizu, Shoichi; Fujiyoshi, Hironobu

doi:10.1145/1959826.1959832

Cited by 8 publications

(3 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First, we need to estimate the 3-D gaze point by using the insideout camera system proposed in [8]. The process of estimating the 3-D gaze point in the camera coordinate system has the following flow.…”

Section: -D Gaze Point Estimationmentioning

confidence: 99%

See 1 more Smart Citation

Recovering 3-D gaze scan path and scene structure from inside-out camera

Yuto

Fujiyoshi

2013

Proceedings of the 4th Augmented Human International Conference

Self Cite

View full text Add to dashboard Cite

First-Person Vision (FPV) is a wearable sensor that takes images from a user's visual field and interprets them, with available information about the user's head motion and gaze, through eye tracking [1]. Measuring the 3-D gaze trajectory of a user moving dynamically in 3-D space is interesting for understanding a user's intention and behavior. In this paper, we present a system for recovering 3-D scan path and scene structure in 3-D space on the basis of egomotion computed from an inside-out camera. Experimental results show that the 3-D scan paths of a user moving in complex dynamic environments were recovered.

show abstract

Section: -D Gaze Point Estimationmentioning

confidence: 99%

“…When these head-mounted eye-tracking systems are used, point-of-regard (POR) is generally measured as a point on the image plane. To acquire a 3-D POR, an inside-out camera system was proposed by Shimizu et al [8]. The inside-out camera system can a recover 3-D POR only under static state.…”

Section: Introductionmentioning

confidence: 99%

Recovering 3-D gaze scan path and scene structure from inside-out camera

Yuto

Fujiyoshi

2013

Proceedings of the 4th Augmented Human International Conference

Self Cite

View full text Add to dashboard Cite

show abstract

“…(2) In the mapping approach, a complementary device, such as the Kinect (Microsoft, USA) or another stereovision system with a SLAM algorithm [Paletta et al 2013] is used to map the environment. The eye tracking data is only used to select the target inside the field of view of the said device [Frisoli et al 2012;Atienza and Zelinsky 2003;Shimizu and Fujiyoshi 2011]. Thanks to the accuracy of stereovision systems [Peng 2011;Andersen et al 2012;Microsoft 2016], this method is more suitable for large workspaces.…”

Section: Introductionmentioning

confidence: 99%

Assessment of accuracy for target detection in 3D-space using eye tracking and computer vision

Leroux

Achiche

Raison

2017

Preprint

View full text Add to dashboard Cite

Over the last decade, eye tracking systems have been developed and used in many fields, mostly to identify targets on a screen, i.e. a plane. For novel applications such as the control of robotic devices by the user vision, there is a great interest in developing methods base on eye tracking to identify target points in free three dimensional environments. The objective of this paper is to characterise the accuracy the eye tracking and computer vision combination that was designed recently to overcome many limitations of eye tracking in 3D space. We propose a characterization protocol to assess the behavior of the accuracy of the system over the workspace of a robotic manipulator assistant. Applying this protocol to 33 subjects, we estimated the behavior of the error of the system relatively to the target position on a cylindrical workspace and to the acquisition time. Over our workspace, targets are located on average at 0.84 m and our method shows an accuracy 12.65 times better than the calculation of the 3D point of gaze. With the current accuracy, many potential applications become possible, such as visually controlled robotic assistants in the field of rehabilitation and adaptation engineering.

show abstract