MirrorBlender: Supporting Hybrid Meetings with a Malleable Video-Conferencing System

In hybrid meetings, multiple co-located participants communicate with remote participants through video. But video communication inhibits non-verbal cues, and this often causes remote participants to feel excluded. To address this issue, we built MirrorBlender: a What-You-See-Is-What-I-See video-conferencing system for blending, repositioning, and resizing mirrors. Mirrors here denote shared video feeds of people and screens. In a qualitative study of MirrorBlender with three hybrid meeting sessions, we found that the shared control of mirrors supported users in negotiating a blended interpersonal space. Moreover, it enabled diverse acts of inclusion of remote participants. In particular, remote participants brought attention to themselves by manipulating the position, scale, and translucency of their camera and screen feeds. Participants also embodied and leveraged their mirror images for deictic gestures and playful interactions. Based on these findings, we discuss new opportunities for supporting video-mediated collaboration.