Smart Mirror Make-up Assistance Algorithm
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Convolutional smart mirrors lack functionality for make-up assistance, failing to provide users with personalized make-up guidance and modification recommendations.
Innovation Solution
A make-up assistance method and apparatus that acquires a facial image, determines differences between the user's make-up and a selected make-up plan using algorithms, and generates prompt information for modifications, allowing real-time adjustments to achieve the desired make-up effect.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If convolutional smart mirror provides basic information display functions, then the device can show news and road conditions, but it cannot provide make-up assistance functionality
Solution Approach 1:
The smart mirror system is enhanced to perform multiple functions including both basic information display and make-up assistance. The mirror integrates camera modules, processing units, and display capabilities to simultaneously provide news/road conditions display and real-time make-up effect comparison, allowing one device to serve multiple purposes without requiring separate dedicated devices
Solution Approach 2:
The make-up assistance functionality is segmented into distinct modules: image acquisition module captures facial images, processing module compares actual make-up with reference images using image processing algorithms, and output module provides visual feedback. This segmentation allows the complex make-up assistance feature to be implemented as independent functional blocks that can be added to the existing smart mirror system
2Productivity
If the smart mirror performs real-time image comparison and analysis, then make-up guidance can be provided, but computational load increases
Solution Approach 1:
The system performs partial image comparison by focusing only on relevant facial regions (eyes, lips, cheeks) rather than processing the entire image. The processing unit compares specific make-up areas against corresponding regions in reference images, reducing the overall computational load while still providing effective make-up guidance for the critical areas that matter most to users
Solution Approach 2:
The system uses the smart mirror's own existing hardware resources (camera, processor, display) to perform the image comparison and analysis functions. The mirror leverages its built-in imaging and processing capabilities rather than requiring external computational devices, thereby minimizing additional energy consumption while enabling real-time make-up assistance
3Ease of operation
If the system provides detailed make-up modification prompts, then user guidance improves, but information processing complexity increases
Solution Approach 1:
The system provides differentiated feedback by region, generating specific modification prompts tailored to each facial area (eyes, lips, cheeks) where make-up differences are detected. Each region receives customized guidance based on its specific characteristics and the reference make-up plan, allowing users to focus on precise local adjustments rather than receiving generic overall instructions, thereby simplifying the make-up application process
Data Source
AI summary
A make-up assistance method includes: acquiring a facial image of a user; acquiring a makeup effect image selected by the user; extracting a make-up region from the facial image of the user, and transmitting the makeup effect image and information about the make-up region to a makeup matching server; performing, by the makeup matching server, a skin detection on the make-up region, so as to obtain skin information of the facial image; determining, by the makeup matching server, at least one makeup plan matched with the make-up region according to the skin information and the makeup effect image; presenting the at least one makeup plan; determining a difference between the makeup effect image and the facial image using a preset algorithm; and generating makeup modification prompt information for a region in the makeup effect image or the facial image where the difference is greater than a threshold.