CN-121979967-A - Intelligent cosmetic suggestion recommendation method, system and equipment based on facial blemishes

CN121979967ACN 121979967 ACN121979967 ACN 121979967ACN-121979967-A

Abstract

The invention discloses an intelligent recommendation method, a system and equipment for cosmetic advice based on facial blemishes, and relates to the technical field of image processing, wherein the method comprises the steps of inputting a preprocessed front face image into a facial blemish segmentation model to obtain a front face blemish segmentation result and blemish types; based on the face flaw segmentation result and flaw types, generating a face flaw evaluation result comprising the positions of flaws of various types on the face and the occupied area of flaws of various types, and inputting the preprocessed face image and the corresponding face flaw evaluation result into a preset question-answer BEiT V model to obtain a makeup recommendation result comprising makeup, makeup skills and cosmetics. The invention provides personalized cosmetic advice rapidly.

Inventors

LI HUANYU
ZHU LING
WEI ZIKUN
LV JUNWEI

Assignees

云南云科特色植物提取实验室有限公司
上海贝芙汀科技有限公司
李寰宇

Dates

Publication Date: 20260505
Application Date: 20231024

Claims (9)

1. The intelligent cosmetic suggestion recommending method based on facial blemishes is characterized by comprising the following steps: acquiring a preprocessed front face image; inputting the preprocessed front face image into a facial flaw segmentation model to obtain a front face flaw segmentation result and flaw types; Generating a face flaw evaluation result based on the face flaw segmentation result and the flaw type, wherein the face flaw evaluation result comprises the positions of flaws of various types on the face and the area occupied by the flaws of various types; And inputting the preprocessed front face image and the corresponding facial flaw evaluation result into a preset question-answer BEiTV model to obtain a makeup recommendation result, wherein the makeup recommendation result comprises makeup, makeup skills and cosmetics.
2. The intelligent recommended method of facial blemish-based cosmetic advice of claim 1, wherein obtaining the preprocessed frontal image comprises: Acquiring a front face image to be processed; And cutting out a face region in the front face image to be processed by adopting a face detection interface MEDIAPIPE of Google open source so as to obtain a preprocessed front face image.
3. The intelligent recommended method of facial blemish-based cosmetic advice of claim 1, wherein the facial blemish segmentation model is obtained by training SwinTransformer neural networks using a first training sample set; The samples in the first training sample set comprise facial sample images with facial blemishes and corresponding facial blemish labels, wherein the facial blemish labels comprise blemish areas and blemish types.
4. The intelligent recommended method of cosmetic advice based on facial blemishes of claim 1 wherein the blemishes types include blemishes, acne marks, sensitive redness, mottle and enlarged pores.
5. The intelligent recommended method for cosmetic advice based on facial blemishes according to claim 1, wherein the text structure of the facial blemish assessment result is the position of blemishes on the face + blemish type + area occupied by blemishes, and the area occupied by blemishes is a percentage value.
6. The intelligent recommended method of cosmetic advice based on facial blemishes of claim 1 wherein the preset question-answer BEiTV model is obtained by training a multimodal BEiT V model with a second training sample set; the multi-mode BEiTV model adopts MultiwayTransformer as a core structure; The samples in the second training sample set comprise a face flaw sample image, a corresponding face flaw evaluation result and a makeup recommendation label, and the makeup recommendation label comprises makeup, makeup skills and cosmetic categories.
7. The intelligent recommended method for cosmetic advice based on facial blemishes of claim 6, wherein the construction process of the recommended label for cosmetic specifically comprises: Crawling corpus information of an Internet makeup community, wherein the corpus information comprises makeup skills, makeup designs and cosmetic selections; Cleaning and checking the obtained corpus information, and marking various types of flaws, corresponding makeup, makeup skills and cosmetic categories to obtain a makeup recommendation database; And selecting a corresponding cosmetic recommendation label from the cosmetic recommendation database based on the face flaw sample image and the corresponding face flaw evaluation result.
8. Facial blemish-based intelligent cosmetic advice recommendation system, which is characterized in that the system comprises: the front face image acquisition module is used for acquiring the preprocessed front face image; the flaw determination module is used for inputting the preprocessed front face image into a face flaw segmentation model so as to obtain a front face flaw segmentation result and flaw types; The facial flaw evaluation module is used for generating a facial flaw evaluation result based on the front face flaw segmentation result and the flaw type, wherein the facial flaw evaluation result comprises the positions of various types of flaws on the front face and the area occupied by the various types of flaws; and the makeup recommendation module is used for inputting the preprocessed front face image and the corresponding facial flaw evaluation result into a preset question-answer BEiT V model to obtain a makeup recommendation result, wherein the makeup recommendation result comprises makeup, makeup skills and cosmetic categories.
9. An electronic device comprising a memory for storing a computer program and a processor that operates the computer program to cause the electronic device to perform the intelligent facial imperfection based cosmetic recommendation method of any one of claims 1-7.

Description

Intelligent cosmetic suggestion recommendation method, system and equipment based on facial blemishes Technical Field The invention relates to the technical field of image processing, in particular to an intelligent recommendation method, system and equipment for cosmetic suggestions based on facial blemishes. Background In every daily scene, make-up can make people possess better mental state, and more people can select to make up in daily life now, especially for the people that has obvious blemish of face, make-up can cover facial blemish well, but make-up's selection, make-up's skill, cosmetic selection etc. are all puzzled problem, all crucial to the effect of last make-up face. Disclosure of Invention The invention aims to provide an intelligent recommendation method, system and equipment for cosmetic advice based on facial blemishes, and personalized cosmetic advice is rapidly provided. In order to achieve the above object, the present invention provides the following solutions: In a first aspect, the present invention provides an intelligent cosmetic advice recommendation method based on facial blemishes, including: acquiring a preprocessed front face image; inputting the preprocessed front face image into a facial flaw segmentation model to obtain a front face flaw segmentation result and flaw types; Generating a face flaw evaluation result based on the face flaw segmentation result and the flaw type, wherein the face flaw evaluation result comprises the positions of flaws of various types on the face and the area occupied by the flaws of various types; And inputting the preprocessed front face image and the corresponding facial flaw evaluation result into a preset question-answer BEiTV model to obtain a makeup recommendation result, wherein the makeup recommendation result comprises makeup, makeup skills and cosmetics. In a second aspect, the present invention provides an intelligent recommended system for cosmetic advice based on facial blemishes, comprising: the front face image acquisition module is used for acquiring the preprocessed front face image; the flaw determination module is used for inputting the preprocessed front face image into a face flaw segmentation model so as to obtain a front face flaw segmentation result and flaw types; The facial flaw evaluation module is used for generating a facial flaw evaluation result based on the front face flaw segmentation result and the flaw type, wherein the facial flaw evaluation result comprises the positions of various types of flaws on the front face and the area occupied by the various types of flaws; and the makeup recommendation module is used for inputting the preprocessed front face image and the corresponding facial flaw evaluation result into a preset question-answer BEiT V model to obtain a makeup recommendation result, wherein the makeup recommendation result comprises makeup, makeup skills and cosmetic categories. In a third aspect, the present invention provides an electronic device, including a memory and a processor, where the memory is configured to store a computer program, and the processor is configured to execute the computer program to cause the electronic device to perform an intelligent recommended method of cosmetic advice based on facial imperfections. According to the specific embodiment provided by the invention, the invention discloses the following technical effects: The invention discloses an intelligent recommendation method, a system and equipment for makeup advice based on facial blemishes, which are used for inputting a preprocessed front face image into a facial blemish segmentation model to obtain a front face blemish segmentation result and blemish types, and further generating a facial blemish assessment result, wherein the facial blemish assessment result comprises positions of various blemishes on a front face and areas occupied by various blemishes, namely, a structured facial assessment conclusion is generated by identifying facial skin blemish conditions. And then inputting the preprocessed front face image and the corresponding facial flaw evaluation result into a preset question-answer BEiT V model to obtain a makeup recommendation result, including makeup, makeup skills and cosmetics. The invention can identify various blemish problems of the face of the user, can grasp mass related knowledge of makeup through the preset question-answering BEiTV model and quickly find out the makeup recommendation result suitable for individuals, specifically combines the face image and the face blemish segmentation result, and outputs personalized makeup advice through the preset question-answering BEiTV model so as to help the user solve the related trouble of makeup. Drawings In order to more clearly illustrate the embodiments of the present invention or the technical solutions of the prior art, the drawings that are needed in the embodiments will be briefly described below, it being obvious that the drawings in