Dynamic Image Text Localization via Layout Parameter Overlay

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for generating image text require designers to manually select and translate layouts for multiple languages, making it inefficient to update textual information across different languages.

Innovation Solution

A dynamic synchronized image text localization method that translates and overlays textual information on images based on determined layout parameters, including directionality, orientation, font type, color, and size, allowing for adaptive changes in response to modifications.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If a designer manually selects and translates layouts for multiple languages, then language-specific layout accuracy is improved, but design time and productivity deteriorate

Engineering Contradiction:
Improvelayout accuracyVSAvoiddesign speed
Core Design Contradiction:
Manufacturing precisionVSProductivity

Solution Approach 1:

The system pre-determines and stores layout parameters for multiple languages before actual translation needs arise. When text needs translation, the pre-computed layout parameters are directly applied, eliminating the need for manual layout selection and enabling rapid multi-language translation while maintaining language-specific layout accuracy

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system automatically adjusts layout parameters (such as text direction, font size, line spacing) based on the target language characteristics. By dynamically changing these parameters according to language requirements, the system achieves accurate language-specific layouts without manual intervention, resolving the contradiction between precision and productivity

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If a designer updates textual information for each language separately, then translation accuracy is improved, but updating time and complexity deteriorate

Engineering Contradiction:
Improvetranslation accuracyVSAvoidupdating time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system uses a universal template structure that can accommodate multiple languages simultaneously. When text is updated in one language, the same update automatically applies to all other languages using the same template, maintaining translation accuracy while eliminating repetitive updating work across different languages

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system merges the updating process for multiple languages into a single operation. By combining all language versions into one synchronized update mechanism, the system achieves accurate translations for all languages while dramatically reducing the time and effort required compared to separate updating processes

Inventive Principle:
Principle #5Merging (Combining)

3Adaptability or versatility

If layout parameters are manually adjusted for each translation, then layout adaptability to language characteristics is improved, but process complexity deteriorate

Engineering Contradiction:
Improvelayout adaptabilityVSAvoidprocess complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system automatically determines and applies appropriate layout parameters based on the target language characteristics without requiring manual adjustment. The system self-adapts to different language requirements (such as right-to-left vs left-to-right text direction) by automatically selecting and applying the correct layout parameters, thereby achieving high layout adaptability while keeping the process simple

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS11347381B2Dynamic synchronized image text localization
Publication Date: 2022.05.31 INTERNATIONAL BUSINESS MACHINE CORPORATION
  • US11347381B2 patent drawing
  • US11347381B2 patent drawing
  • US11347381B2 patent drawing

AI summary

A set of images and a corresponding set of textual information are received, such that a first textual information is to be presented with a first image. The first textual information is translated into a set of translated textual information, each textual information in the set of translated textual information corresponding to a language in a set of languages. A set of layout parameters is determined for at least one translated textual information. The translated textual information is transformed according to the set of layout parameters and overlaid on the first image. The overlaying causes a pixel of the transformed translated textual information to be superimposed on a pixel of the first image.