Eye Makeup Transformation via Triangulation Mesh Key Points

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional image processing methods for eye makeup struggle to dynamically adjust to changing eye states, leading to distortion and a poor user experience due to differences in eyebrow shapes and eye positions between individuals.

Innovation Solution

An image processing method that recognizes eye key points, performs interpolation to obtain auxiliary key points, and transforms an eye makeup effect image based on a triangulation mesh, ensuring similarity between corresponding triangles in the source and target images within a preset error range, thereby maintaining the shape of the triangulation mesh regardless of eye state changes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If conventional face image mesh methods with triangulation are used, then the whole image can be divided into multiple triangle regions, but the triangulation mesh shape varies greatly depending on different eye states and individual differences, leading to distortion

Engineering Contradiction:

Improveadaptability to different eye statesVSAvoidtriangulation mesh shape consistency

Core Design Contradiction:

Adaptability or versatilityVSManufacturing precision

Solution Approach 1:

The patent applies preliminary action by pre-defining a standard eye template with fixed key points and triangulation mesh structure. Before processing individual eye images, the system prepares this standard template in advance, which serves as a reference for subsequent matching and transformation operations, ensuring consistent mesh shape across different eye states

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent employs parameter changes by dynamically adjusting transformation parameters (translation, rotation, scaling) based on the detected eye key points and state. Instead of changing the mesh structure itself, the system modifies transformation parameters to adapt the standard template to different eye states while maintaining mesh shape consistency

Inventive Principle:

Principle #35Parameter changes

2Ease of operation

If the triangulation mesh is adjusted to fit different eye positions, then the eye makeup effect can be applied to various positions, but the mesh shape distortion occurs due to individual differences in eyebrow shapes and eye positions

Engineering Contradiction:

Improveeye makeup application flexibilityVSAvoidtriangulation mesh shape integrity

Core Design Contradiction:

Ease of operationVSShape

Solution Approach 1:

The patent applies copying by creating a standardized eye template that contains the ideal triangulation mesh structure. This template is copied and applied to different user eyes through transformation operations rather than modifying the original mesh structure, thus maintaining shape integrity while achieving positioning flexibility

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent employs dynamics by implementing dynamic transformation parameters that adjust the position, scale, and rotation of the standard eye template based on detected eye key points. This allows the makeup effect to dynamically adapt to different eye positions and states while the underlying mesh structure remains consistent

Inventive Principle:

Principle #15Dynamics

3Measurement precision

If eye key points are recognized and interpolation is performed to obtain auxiliary key points, then the triangulation mesh can be formed at predetermined eye position, but the processing complexity increases

Engineering Contradiction:

Improveeye key point detection accuracyVSAvoidimage processing complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies segmentation by dividing the eye region into specific key point locations (eyebrow endpoints, eye corners, etc.) that need to be detected. By segmenting the detection task into discrete key points rather than analyzing the entire eye region, the system achieves high precision while managing processing complexity through focused detection

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS11017580B2Face image processing based on key point detection

Publication Date: 2021.05.25 BEIJING MICROLIVE VISION TECH CO LTD

AI summary

There are provided an image processing method, an image processing device, a hardware device for image processing, a computer readable storage medium, and an image processing terminal. The image processing method includes: recognizing eye key points on a face image; performing interpolation to obtain auxiliary key points, where the auxiliary key points and the eye key points form a first triangulation mesh at a predetermined eye position on the face image; and transforming, based on the first triangulation mesh, an eye makeup effect image to the predetermined eye position.