Facial Ratio Image Correction via Feature Point Manipulation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional image correction techniques fail to generate filters that suit individual users' facial features in real time, leading to inconvenient and unsuitable corrections, as they rely on pre-generated filters based on specific variations rather than personalized facial ratios.

Innovation Solution

An image correction method that allows users to directly move facial feature points on a screen to generate correction pattern information based on changed facial ratios, enabling real-time image correction and automatic adaptation for all users, with the option to optimize and update this information for improved usability.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If pre-generated filters based on specific variations are used, then the filtering process is simplified, but the filters cannot be generated using the user's actual face and are not suitable for all users

Engineering Contradiction:

Improvefilter generation processVSAvoidfilter suitability for different users

Core Design Contradiction:

Ease of manufactureVSAdaptability or versatility

Solution Approach 1:

The system performs preliminary action by pre-generating multiple filters based on different shape styles (e.g., big eye, high nose) before the user needs them. These filters are stored and can be quickly applied without real-time generation, simplifying the manufacturing process while maintaining adaptability through user selection.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system introduces dynamics by allowing users to select and apply different pre-generated filters based on their desired shape styles. The filter application process dynamically adjusts the input image based on the selected filter, enabling adaptability to different user preferences without requiring real-time filter generation.

Inventive Principle:

Principle #15Dynamics

2Adaptability or versatility

If users manually re-generate filters to suit new shape styles, then the filters can be customized, but the process becomes complex and inconvenient

Engineering Contradiction:

Improvefilter customizationVSAvoidfilter generation operation

Core Design Contradiction:

Adaptability or versatilityVSEase of operation

Solution Approach 1:

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system uses copying by creating multiple pre-generated filter versions based on different shape styles. Instead of requiring users to manually generate custom filters, the system provides copied versions of popular shape style filters that can be directly applied, significantly reducing operational complexity.

Inventive Principle:

Principle #26Copying

3Device complexity

If filters only cause distortion by variations for a specific region, then the filter generation is simple, but it is difficult to make corrections that suit all users

Engineering Contradiction:

Improvefilter generation complexityVSAvoidcorrection suitability for all users

Core Design Contradiction:

Device complexityVSAdaptability or versatility

Solution Approach 1:

The system achieves universality by creating filters that can be applied to all users regardless of their specific facial features. The pre-generated filters are designed to work universally across different face types, and the system provides multiple filter options to accommodate various user needs, making the correction process suitable for all users without requiring complex individualized generation.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS11281890B2Method, system, and computer-readable media for image correction via facial ratio

Publication Date: 2022.03.22 SNOW CORP

AI summary

Provided is a non-transitory computer-readable medium storing instructions for performing image correction by determining a plurality of first facial feature points of a first face included in a first image, displaying (i) at least a portion of the plurality of first facial feature points and (ii) the first image on a screen of an electronic device, recognizing a first input from an external source for moving at least one first facial feature point among the plurality of first facial feature points in the first image displayed on the screen, generating a corrected first image by moving the at least one first facial feature point in response to the first input and correcting the first image based on the moved at least one first facial feature point, and generating correction pattern information by analyzing a correction pattern of the first image.