ARTIFICIAL INTELLIGENCE VIRTUAL MAKEUP METHOD AND DEVICE USING MULTI-ANGLE IMAGE RECOGNITION

Smart overview of the Invention

The patent application outlines a method and device for virtual makeup application using artificial intelligence and multi-angle image recognition. The process involves collecting a source image, transforming it through face warping based on a benchmark image, and constructing a style database. This database maps style information from the source image, enabling a personalized style recommendation service. The system performs style transfer by face warping the user’s image, applying style transfer, and reverse warping to produce a virtual style synthesis image.

Technical Field and Background

This invention pertains to the beauty industry, specifically addressing the growing demand for personalized makeup styles that consider individual variations in skin color, facial shape, and features. Traditional virtual makeup applications lack the ability to offer personalized recommendations and rely on limited data, often excluding expert collaboration. The proposed technology leverages machine learning and deep learning to enhance makeup evaluation and style recommendation services.

Method and Device Functionality

The system includes a face warping processor to transform source images into warping images, constructing a style database. A service processor then offers style recommendations through style transfer. This involves face warping a user’s image, conducting style transfer, and reverse warping to create a virtual style synthesis image. The system adjusts the transfer intensity for each image area, providing a comprehensive style recommendation service.

Technical Effects and User Benefits

The device and method improve user convenience by providing accurate and personalized style recommendations. Users can input their face image, and the system recommends synthesis images based on virtual style synthesis and face analysis. This enables users to easily select suitable makeup styles, enhancing satisfaction with the service’s accuracy and customization.

Illustrations and Embodiments

The application includes diagrams and flowcharts illustrating the system's structure and process. Figures depict the overall system, detailed service providing device, face warping processing, transfer intensity adjustment, and face analysis units. Additional diagrams show facial partial age estimation and style interface outputs, aiding in understanding the device's operation and potential applications.