Recently, live video streaming from hand- held devices entered mass market . Ordinary video conference systems were supplemented with additional features, such as marking in order to point attention of the conference participants on specific object in the video. However, the accuracy of live video marking depends on the video latency, received between two video conference participants.
The aim of this study was to propose a solution to synchronize video object coordinates in two video streams: transmitted and received with latency that is close to 2 seconds. A new system was proposed in this paper designed to track an object in the video stream based on the inertial sensor data. It was found that the displacement of the object of interest during latency interval could be predicted by the use of inertial sensors of the handheld device with 86% accuracy in average.