Translucent Display Video Communication Eye Contact Alignment

The patent uses multiple cameras to capture video feeds from different angles and positions, then synthesizes these feeds to create a virtual image that appears at the correct position on the display. This copying approach allows the system to simulate eye contact by presenting a processed version of the video that aligns the remote user's eyes with the local user's camera position, resolving the contradiction between separate component placement and eye contact alignment.

The patent introduces a video processing system as an intermediary between the camera array and the display. This intermediary processes multiple video feeds, adjusts their positions and orientations, and synthesizes them into a composite image that creates the illusion of direct eye contact. The intermediary reconciles the physical separation of camera and display components with the visual requirement for eye alignment.

The patent divides the video capture function into multiple segments, with each camera responsible for capturing video from a specific angle or position. This segmentation allows the system to gather comprehensive visual data from different perspectives, which is then processed to create the aligned eye contact effect. By breaking down the complex task of capturing multi-angle video into separate camera functions, the system manages device complexity through functional division.

The patent makes each camera in the array multi-functional by using them not only for capturing video feeds but also for determining spatial relationships and user positions. The same camera hardware serves multiple purposes: video capture, position detection, and alignment reference. This universality reduces the need for separate specialized components, thereby managing device complexity while achieving eye contact alignment.

The patent implements a feedback mechanism where the system continuously monitors the positions of users relative to the camera and display, then adjusts the video feed positioning accordingly. The video processing system receives feedback about user locations and uses this information to dynamically adjust the alignment of video feeds from multiple cameras, ensuring that the synthesized image maintains accurate eye contact alignment while capturing the correct user image positions.

The patent employs dynamic adjustment of video feed positions and orientations based on real-time user positions. Rather than using fixed camera-display alignment, the system dynamically recalibrates the virtual image positioning to account for user movement and changing spatial relationships. This dynamic approach maintains both eye contact alignment and accurate user image capture despite position variations.