AR Makeup Simulation with 3D Facial Feature Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current makeup simulation technologies struggle to provide real-time, multi-angle views of makeup applications and do not effectively assist users in applying makeup based on simulated results.
Innovation Solution
An information processing apparatus using augmented reality (AR) technology to capture facial images, recognize facial features, and superimpose virtual makeup images onto real-time video feeds, allowing users to see makeup application procedures and completion states from various angles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a synthesized image generated by makeup simulation is displayed as a still image from a front face, then the makeup simulation can be performed, but it is difficult to confirm makeup simulation from several angles in real time
Solution Approach 1:
The patent transitions from 2D still images to 3D virtual models that can be viewed from multiple angles. The 3D virtual face model allows users to rotate and view makeup applications from front, side, and other perspectives, adding spatial dimensionality to the makeup simulation experience.
Solution Approach 2:
The system transforms static synthesized images into dynamic real-time video feeds. The virtual makeup images are applied to moving video of the user's face, allowing continuous updates and real-time interaction as the user moves their head or applies makeup, making the simulation dynamic rather than static.
2Ease of operation
If a synthesized image generated by makeup simulation is displayed, then the state after makeup completion can be confirmed, but it is difficult to enter a simulated state by a user actually doing makeup
Solution Approach 1:
The system provides real-time feedback by overlaying virtual makeup images on the user's actual face video feed. As the user applies makeup, the system continuously compares their actions with the simulated makeup application process and provides visual guidance, allowing the user to see both their progress and the target result simultaneously.
Solution Approach 2:
The system performs preliminary makeup simulation to generate target makeup images before the user actually applies makeup. This pre-simulation shows the user what the final result should look like, providing a reference guide for the actual makeup application process.
3Speed
If traditional makeup simulation methods are used, then a synthesized image can be generated, but real-time visualization of makeup application procedures from various angles is not possible
Solution Approach 1:
The system segments the facial image into multiple feature points (eyes, eyebrows, nose, lips, etc.) and processes each segment independently. This allows parallel processing of different facial regions, improving real-time performance while maintaining accurate recognition of each facial feature's position and characteristics.
Data Source
AI summary
A method is provided for generating output image data. The method comprises receiving image data representing an input image, the input image containing at least one facial image. The method further comprises recognizing the facial image in the image data, and recognizing facial features of the facial image. The method further comprises generating data representing a makeup image based on the recognized facial features, the makeup image providing information assisting in the application of makeup. The method also comprises generating output image data representing the makeup image superimposed on the facial image.