Speech Image Overlay Positioning for 3D Target Recognition

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing systems fail to intuitively recognize which target is being supplemented by supplementary images in speech images, particularly when these images are displayed at two-dimensional positions without considering the three-dimensional context.

Innovation Solution

An information processing system that acquires a speech image including a speaking person, generates a supplementary image based on spoken content, and controls its display at a two-dimensional position corresponding to the three-dimensional position of the target in the speech image, using techniques such as sign language interpretation or object recognition to enhance intuitive recognition.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If supplementary images are displayed at two-dimensional positions without considering three-dimensional position, then device complexity is reduced, but recognition intuitiveness deteriorates

Engineering Contradiction:
Improvedisplay control complexityVSAvoidrecognition intuitiveness
Core Design Contradiction:
Device complexityVSLoss of information

Solution Approach 1:

The patent applies dimensionality change by mapping three-dimensional spatial position information onto two-dimensional display positions. The display control unit calculates two-dimensional coordinates based on three-dimensional position data, thereby preserving spatial relationship information while adapting to the two-dimensional display medium. This resolves the contradiction by maintaining recognition intuitiveness through spatial correspondence without requiring complex three-dimensional display hardware.

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

2Ease of operation

If supplementary images are displayed at two-dimensional positions without considering three-dimensional position, then ease of operation is improved, but information accuracy deteriorates

Engineering Contradiction:
Improvedisplay implementation easeVSAvoidposition correspondence accuracy
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The patent transforms three-dimensional position information into two-dimensional display coordinates while maintaining spatial relationship accuracy. This approach simplifies display implementation on two-dimensional screens while preserving the accuracy of position correspondence between speech images and supplementary images through mathematical coordinate transformation.

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

Solution Approach 2:

The display control unit acts as an intermediary that processes three-dimensional position information and converts it into appropriate two-dimensional display positions. This intermediary function maintains position correspondence accuracy by calculating optimal two-dimensional coordinates that reflect the original three-dimensional spatial relationships, thereby ensuring information accuracy while simplifying display operations.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP4664245A1Information processing system, program, and information processing method
Publication Date: 2025.12.17 FUJIFILM BUSINESS INNOVATION CORP
  • EP4664245A1 patent drawingFigure 1
  • EP4664245A1 patent drawingFigure 2
  • EP4664245A1 patent drawingFigure 3

AI summary

An information processing system includes one or plural processors configured to acquire a speech image including a speaking person, acquire a supplementary image for supplementing spoken content of the speaking person, the supplementary image being based on the spoken content and not being a text directly representing the spoken content, and perform a control of displaying the supplementary image at a two-dimensional position in the speech image corresponding to a three-dimensional position at which a target to be supplemented is present in the speech image.