Avatar Video Rendering for Multi-User Participation Load Control

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing information processing systems that distribute videos with avatar objects struggle to increase participation opportunities among users, limiting the number of participants and viewing time.

Innovation Solution

Implement a program that outputs a first avatar object with tracking data and, upon a predetermined user action, displays a second avatar object with reduced rendering load and corresponding information within a shorter display period, allowing multiple users to participate in video distribution.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If multiple users participate in video distribution with full rendering, then user participation opportunities increase, but rendering processing load increases excessively

Engineering Contradiction:

Improveuser participation opportunitiesVSAvoidrendering processing load

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent applies different rendering qualities to different user avatar objects based on their roles and positions in the video. The first user's avatar object receives full rendering processing, while second users' avatar objects receive reduced rendering processing. This local differentiation allows multiple users to participate without uniformly increasing the rendering load for all participants, thus resolving the contradiction between user participation opportunities and rendering processing load.

Inventive Principle:

Principle #3Local quality

2Reliability

If second avatar objects are displayed with full rendering, then participation visibility improves, but processing time and load increase

Engineering Contradiction:

Improveparticipation visibilityVSAvoidprocessing time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent implements partial rendering action for second avatar objects by displaying only essential information (such as avatar objects and text information) without applying full rendering processing. This partial action approach maintains sufficient visibility for participant identification while significantly reducing the processing time required, thus resolving the contradiction between participation visibility and processing time.

Inventive Principle:

Principle #16Partial or excessive action

3Productivity

If multiple avatar objects are displayed simultaneously, then user engagement increases, but system resource consumption increases

Engineering Contradiction:

Improveuser engagementVSAvoidsystem resource consumption

Core Design Contradiction:

ProductivityVSUse of energy by moving object

Solution Approach 1:

The patent changes the rendering parameters for different avatar objects based on their importance and position. The first avatar object uses standard rendering parameters, while second avatar objects use reduced rendering parameters (such as lower resolution, simplified models, or reduced frame rates). This parameter differentiation allows multiple avatar objects to be displayed simultaneously, increasing user engagement, while controlling overall system resource consumption by allocating resources differently across objects.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS20260046473A1Program, information processing method, and information processing device

Publication Date: 2026.02.12 GREE INC

AI summary

A user device executes (i) a first output control process that outputs, to a display, a video including a first avatar object to which tracking data showing movement of a distributing user is applied; and (ii) a second output control process that, when an action with respect to the video of a second user who views the video corresponds to a predetermined reference action, outputs to the display a video including (a) a second avatar object to which movement of the second user is not applied, and (b) the first avatar object.