CNIPA.AI
검색으로 돌아가기
기록

METHOD AND DEVICE FOR DETECTING FACIAL WRINKLES USING DEEP LEARNING-BASED WRINKLE DETECTION MODEL TRAINED ACCORDING TO SEMI-AUTOMATIC LABELING

발명심사 중
6청구항 · 1 독립항
§ Ⅰ

개요

발명자

Semin KIM; Huisu YOON; Jong Ha LEE; Sangwook YOO; Yongjoon CHOE

IPC 분류

G6V 40/16G6V 10/30G6V 10/44G6V 10/54G6V 10/764G6V 10/77G6V 10/776

CPC 분류

G6V40/172G6V10/30G6V10/443G6V10/54G6V10/764G6V10/7715G6V10/776G6V40/171

Disclosed are a method and device for detecting facial wrinkles using a deep learning-based wrinkle detection model trained according to semi-automatic labeling. The method for detecting facial wrinkles using a deep learning-based wrinkle detection model trained according to semi-automatic labeling comprises the steps of: generating labeling data; using the generated labeling data to train a wrinkle detection model using supervised learning; inputting a user's face image to the wrinkle detection model trained using supervised learning; and obtaining wrinkle detection data corresponding to the face image on the basis of the output of the wrinkle detection model. Therefore, wrinkles on the face can be detected by quickly and accurately obtaining labeling data by generating the labeling data semi-automatically.

원문 (중국어)

Disclosed are a method and device for detecting facial wrinkles using a deep learning-based wrinkle detection model trained according to semi-automatic labeling. The method for detecting facial wrinkles using a deep learning-based wrinkle detection model trained according to semi-automatic labeling comprises the steps of: generating labeling data; using the generated labeling data to train a wrinkle detection model using supervised learning; inputting a user's face image to the wrinkle detection model trained using supervised learning; and obtaining wrinkle detection data corresponding to the face image on the basis of the output of the wrinkle detection model. Therefore, wrinkles on the face can be detected by quickly and accurately obtaining labeling data by generating the labeling data semi-automatically.