When a participant in a video conference looks to address another participant, eye-to-eye contact is not made. This is because neither participant is looking at their respective camera; rather, each participant looks at their display screen, towards the video feed of the other participant, which is often displayed at a position that is some distance away from the device camera. This is a major cognitive difference of video conferencing from real-life meeting of people. This disclosure describes techniques that, with user permission, adjust eye gaze direction in video conferencing by (a) capturing a video feed using two or more cameras, e.g., on the left and right side of a user device; (b) building a real-time 3D model of the participant’s face; and (c) by re-rendering the face as if a virtual camera was right in between the two cameras. The described techniques can provide the experience of direct eye contact during a video conference.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.