Composite Telepresence Image Merging for Screen Space Utilization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current video conferencing systems, including consumer telepresence systems, face the challenge of displaying multiple remote participant images in a way that maximizes screen space, as separate windows lead to wasted space and participants appear smaller than life-size due to tiling issues.

Innovation Solution

A method is implemented at the local station to process and merge participant images from remote stations, isolating participants from their backgrounds and overlaying them to create a composite image that can be displayed on a single monitor, allowing for realistic and life-sized representation of participants.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Area of moving object

If multiple remote participant images are displayed in separate windows, then each participant can be clearly displayed, but screen space is wasted and participants appear smaller than life-size

Engineering Contradiction:
Improvescreen space utilizationVSAvoidparticipant image size accuracy
Core Design Contradiction:
Area of moving objectVSManufacturing precision

Solution Approach 1:

The patent merges multiple participant images into a single composite display by isolating each participant from their background and layering them together. This combining approach eliminates the need for separate windows, maximizes screen space utilization, and allows participants to be displayed at life-size proportions through coordinated scaling and positioning operations.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent extracts each participant image from its original background context using background subtraction techniques. By removing the background and isolating only the participant, the system can then reposition and resize these extracted images freely within the composite display, achieving life-size representation without the constraints of separate window layouts.

Inventive Principle:
Principle #2Taking out (Extraction)

2Manufacturing precision

If a room-sized screen is used to display life-sized participants, then participants appear life-sized, but the system becomes impractical for consumer use

Engineering Contradiction:
Improveparticipant image size accuracyVSAvoiddisplay system complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent transitions from a spatial solution (room-sized screens) to a computational solution by introducing multiple visual dimensions: layering, transparency, and positioning. Through coordinate transformations and scaling operations, the system creates a composite image space where multiple life-sized participants can coexist on a standard screen, effectively adding dimensional complexity to the display approach.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Area of moving object

If participant images are isolated from background and merged, then screen space is maximized and life-sized display is achieved, but image processing complexity increases

Engineering Contradiction:
Improvescreen space utilizationVSAvoidimage processing complexity
Core Design Contradiction:
Area of moving objectVSDevice complexity

Solution Approach 1:

The patent performs preliminary background isolation and participant extraction before the composite display assembly. By pre-processing each participant image to remove backgrounds and isolate subjects, the system simplifies the subsequent merging process, reducing real-time processing complexity while achieving the desired life-sized composite display effect.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS9756288B2Tiering and manipulation of peer's heads in a telepresence system
Publication Date: 2017.09.05 INTERDIGITAL CE PATENT HOLDINGS SAS
  • US9756288B2 patent drawing
  • US9756288B2 patent drawing
  • US9756288B2 patent drawing

AI summary

In a telepresence system, a local station receives images of participants at remote stations processed to isolate the remote participant in the image from image background. The local station merges the images of the participants at the remote stations to generate an output image in which a first remote participant appears to partially overlay another participant. The output image undergoes display at the local station.