Context-aware automatic video screen manipulation using trajectory tracking

Course project in KAIST Mobile Computing and Systems for Intelligent Living course (EE595)

Project Summary

Numerous people watch videos through monitor or smartphone screen. We point out the limitations of automatic screen manipulation according to the viewer's physical context when watching video through monitor. And to provide a better experience for video viewers, we present a service with context-aware 1) automatic screen size adjustment, 2) automatic screen rotation, and 3) automatic video switch between devices. We first show the improvement in trajectory tracking through hardware adjustment and integration error reduction, which leads to the improvement of the performance of our whole system. Then we describe the architecture of our system, consisting of integration of trajectory tracking, python client, web client, and mobile client. The working system is demonstrated through the linked video in the report.

Libraries & Frameworks

  • Arduino
  • Android
  • Firebase
Master’s student @ KAIST EE

Seungjoo Lee