Artificial intelligence virtual makeup method and device using multi-angle image recognition

A service providing device comprising:

• a face warping processor configured to collect a source image and process face warping of the collected source image to transform the collected source image into a warping image based on a benchmark image;
• a style database configured to construct a style database to which style information of the source image is mapped, using the face warped source image;
• and
• a service providing processor configured to provide a style recommendation service based on style transfer using the style database, wherein
• the service providing processor performs face warping on a user image input from a user terminal, performs style transfer of the face warped source image in response to the face warped user image, and performs face reverse warping of the image subjected to the style transfer to output as a virtual style synthesis image, and
• the processing of the face warping of the collected source image to transform the collected source image into the warping image based on the benchmark image includes:
• determining the benchmark image;
• identifying landmark points of the benchmark image;
• performing triangulation corresponding to the landmark points;
• storing one or more pieces of triangle information determined according to the performed triangulation;
• extracting a partial area image corresponding to first triangle information of the benchmark image from the source image;
• and
• calculating a transformation matrix for transforming triangle information extracted from the partial area image into the first triangle information.

A method of operating a service providing device, the method comprising:

• collecting a source image and processing face warping of the collected source image to transform the collected source image into a warping image based on a benchmark image;
• constructing, using the face warped source image, a style database to which style information of the source image is mapped;
• and
• providing a style recommendation service based on style transfer using the style database,
• wherein the processing of the face warping includes:
• determining the benchmark image;
• identifying landmark points of the benchmark image;
• performing triangulation corresponding to the landmark points;
• and
• storing one or more pieces of triangle information determined according to the performed triangulation, and
• wherein the processing of the face warping further includes:
• extracting a partial area image corresponding to first triangle information of the benchmark image from the source image;
• and
• calculating a transformation matrix for transforming triangle information extracted from the partial area image into the first triangle information.