Face image portrait-buffing method and device thereof, computer equipment and storage medium

Abstract

The invention discloses a face image portrait-buffing method comprising the steps that face key points in an original face image are recognized, a to-be-buffed face area is determined according to therecognized face key points, and the to-be-buffed face area is a face skin area; gaussian blur processing is carried out on the face area to be subjected to portrait-buffing to obtain a buffed face area image; and superposition processing is carried out on the face region image after portrait-buffing and the original face image to obtain a face image after portrait-buffing. In the whole process, the face area to be subjected to portrait-buffing is determined through the face key points, the skin area recognition accuracy is improved, detail characteristics of a non-skin area of the face imageafter the portrait-buffing cannot be lost, and the image after the portrait-buffing is more natural. In addition, the invention further provides a face image portrait-buffing device, computer equipment and a storage medium.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a face image skin smoothing method, device, computer equipment and storage medium. Background technique [0002] With the popularity of beautification, more and more users hope to beautify the characters in pictures or videos through beautifying tools. Especially in the field of mobile phones or live broadcasts, the demand for beauty is becoming more and more prominent. Among them, microdermabrasion is the most important part of beauty. Generally, a method of identifying skin color is used to determine the skin area. However, skin color recognition is very sensitive to light. When the light is dark, the skin area cannot be fully recognized. When the light is strong, the range of the skin area will be expanded, making the user's face skin color uneven or facial contours in photos or videos. Vague. [0003] Currently, the mainstream dermabrasion methods on the market ...

AI Technical Summary

Problems solved by technology

However, skin color recognition is very sensitive to light. When the light is dark, the skin area cannot be fully recognized. When the light is strong, the range of the skin area will be expanded, making the user's face skin color uneven or facial contours in photos or videos. Vague

[0003] At present, the mainstream skin microdermabrasion methods on the market all have the disadvantage of losing the image details of non-skin areas. For example, the microdermabrasion treatment around the eyes will make the details inside the eyes blurred, thus affecting the overall clarity of the face image.

Images