Egocentric Video Multi-viewer for Analyzing Skilled Behaviors based on Gaze Object

In many intellectual tasks, efficient succession of human physical and sensory skills is a long-standing issue. In order to analyze skilled behaviors, a useful approach is to compare same task scenes among workers or days and to understand these differences. In this paper, we propose an egocentric scene classification method based on objects which the worker turned the gaze to and a multi-viewer for egocentric videos comparison. We have experimented on proposed classification method with videos of painting watercolor.

[1]  A. L. I︠A︡rbus.  Eye Movements and Vision , 1967 .

[2]  Ali Farhadi,et al.  You Only Look Once: Unified, Real-Time Object Detection , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[3]  Takatsugu Hirayama,et al.  Skilled gaze behavior extraction based on dependency analysis of gaze patterns on video scenes , 2016, ETRA.

[4]  B. Tatler,et al.  Yarbus, eye movements, and vision , 2010, i-Perception.

[5]  A. L. Yarbus,et al.  Eye Movements and Vision , 1967, Springer US.