CamCutter: Impromptu Vision-Based Cross-Device Application Sharing

Takuma Hagiwara,Kazuki Takashima,Morten Fjeld,Yoshifumi Kitamura
DOI: https://doi.org/10.1093/iwcomp/iwz035
2019-11-01
Interacting with Computers
Abstract:Abstract As the range of handheld, mobile and desktop devices expands and worldwide demand for collaborative application tools increases, there is a growing need for higher speed impromptu cross-device application sharing to keep up with workplace requirements for on-site or remote collaborations. To address this, we have developed CamCutter, a cross-device interaction technique enabling a user to quickly select and share an application running on another screen using the camera of a handheld device. This technique can accurately identify the targeted application on a display using our adapted computer vision algorithm, system architecture and software implementation, allowing impromptu real-time and synchronized application sharing between devices. For desktop and meeting room set-ups, we performed a technical evaluation, measuring accuracy and speed of migration. For a single-user reading task and a collaborative composition task, we carried out a user study comparing our technique with commercial screen sharing applications. The results of this study showed both higher performance and preference for our system. Finally, we discuss CamCutter’s limitations and present insights for future vision-based cross-device application sharing.
computer science, cybernetics,ergonomics
What problem does this paper attempt to address?