Face replacement method, device and electronic equipment

A face replacement and face expression technology, applied in the field of image processing, can solve the problem of unrealistic results, and achieve the effect of avoiding rendering results

Active Publication Date: 2020-05-08
赣州朗顺数字科技有限公司
View PDF12 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] However, due to differences in head shape and face shape from person to person, only replacing the actor's face will make the result unrealistic
Moreover, the facial expression model generally does not include lips. When the actual degree of the actor's mouth opening is inconsistent with the distribution of mouth feature points obtained by the face detection API, the rendering result of "mouth in mouth" may appear

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Face replacement method, device and electronic equipment
  • Face replacement method, device and electronic equipment
  • Face replacement method, device and electronic equipment

Examples

Experimental program
Comparison scheme
Effect test

no. 1 example

[0033] Please refer to figure 2 , figure 2 A flowchart of a face replacement method provided by an embodiment of the present invention is shown. The face replacement method includes the following steps:

[0034] Step S101, acquiring a target frame image of the original video, wherein the target frame image is any frame image of the original video.

[0035] In the embodiment of the present invention, the face replacement method can implement face replacement for specified actors in videos (such as TV dramas, movies, short films, etc.), and can also implement face replacement for specified characters in pictures. As an implementation manner, the original video may be, but not limited to, a TV series, movie, short film and other videos containing specified actors.

[0036] Step S102, in the target frame image, determine the head area of ​​the first person.

[0037] In the embodiment of the present invention, the first person is a person to be replaced for face replacement, ...

Embodiment approach

[0049] In the embodiment of the present invention, the method for matching the head area of ​​the rendered image with the head area of ​​the first person in the target frame image may be: first, detecting feature points of the head area of ​​the rendered image; second, detecting The feature points of the head area of ​​the first person in the target frame image; third, carry out affine transformation on the feature points obtained in the previous two steps, so as to combine the head area of ​​the rendered image with the head area of ​​the first person in the target frame image Fourthly, Gaussian smoothing is performed on the head area of ​​the rendered image according to the head area of ​​the first person in the previous frame / next frame image of the target frame image to obtain the first image. As an implementation manner, the feature points may include feature points of the edge contour of a human face and feature points of facial features of a person.

[0050] Please refer...

no. 2 example

[0074] Please refer to Figure 7 , Figure 7 A schematic block diagram of the face replacement device 200 provided by the embodiment of the present invention is shown. The face replacement device 200 includes a target frame image acquisition module 201 , an area determination module 202 , a background restoration module 203 , a first image acquisition module 204 , an expression tracking module 205 and a second image acquisition module 206 .

[0075] The target frame image acquisition module 201 is configured to acquire a target frame image of the original video, wherein the target frame image is any frame image of the original video.

[0076] In the embodiment of the present invention, the target frame image acquisition module 201 may be used to execute step S101.

[0077] The area determination module 202 is configured to determine the head area of ​​the first person in the target frame image.

[0078] In the embodiment of the present invention, the area determination modu...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The present invention relates to the technical field of image processing, and provides a face replacement method, device and electronic equipment, by determining the head area of ​​a first person in a target frame image, and rendering the 3D head model of a second person to the point where the first person is removed From the background image of the head area of ​​a character, the first image including the head area of ​​the second character and the torso area of ​​the first character can be obtained, which can ensure the complete replacement of the head of the character, and solve the problem of only replacement in the prior art. Faces cause problems with unrealistic results. In addition, by tracking the expression of the first person in the target frame image, the first model including the mixed face model and lip model is obtained, and then the first model is rendered to the head area of ​​the second person in the first image to obtain The second image avoids the rendering result of "mouth in mouth" in the prior art.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to a face replacement method, device and electronic equipment. Background technique [0002] With the development of virtual reality, virtual human face technology (also known as "face changing") has also been greatly developed, for example, it is applied in the replacement of movie roles. At present, a commonly used implementation method of virtual face technology is: first, use face detection technology to obtain face feature points of the target frame and material frame, and make three-dimensional models of different expressions; The relative offset between the feature points and the feature points of the standard face is used to control the degree of deformation of each 3D model and to mix them; finally, the face feature points of the material frame are fused to the mixed model to complete the rendering. [0003] However, due to differences in head shape and fa...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/00G06T13/40G06T15/00
CPCG06T13/40G06T15/005G06V40/176G06V40/168
Inventor 马兆远侯沛宏郝君梁冬雪
Owner 赣州朗顺数字科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap