A face image transformation method based on face 3D reconstruction and face alignment
A face alignment and three-dimensional reconstruction technology, applied in the field of face image transformation, can solve problems such as difficult implementation and inability to use general models, and achieve the effect of ensuring accuracy
Active Publication Date: 2019-03-01
四川云从天府人工智能科技有限公司
View PDF25 Cites 8 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
This requires training the corresponding model for each specific face, instead of using a general model, which is difficult to achieve in many scenarios
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0023] The preferred embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings, so that the advantages and features of the present invention can be more easily understood by those skilled in the art, so as to define the protection scope of the present invention more clearly.
[0024] Such as figure 1 Shown: a face image transformation method based on face 3D reconstruction and face alignment,
[0025] Take the following steps,
[0026] Step 1: the three-dimensional image acquisition device collects a three-dimensional image of a face as an input image, and transmits the input image to a processing module;
[0027] Step 2: the face recognition module extracts the three-dimensional features of the input image;
[0028] The three-dimensional features of the input image include coordinate values of facial feature point sets;
[0029] A standard face alignment feature set is stored in the standard face database, the fa...
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
Login to View More Abstract
The invention relates to a face image transformation method based on face 3D reconstruction and face alignment. The method includes steps: 1, a human face three-dimensional image is collected by a three-dimensional image collecting device as an input image and transmits the input image to a processing module; 2, extracting three-dimensional features of that input image by a face recognition module; the three-dimensional features of the input image include coordinate values of a set of facial feature points; a standard face alignment feature set is stored in the standard face database. The facealignment feature set is composed of face alignment feature points, and each face alignment feature point is the semantic information of the corresponding point on the human face. 3, a reference picture database is arranged, and that processing module extract a reference picture from the reference picture database and transmits the reference picture to the face recognition module. The face imagetransformation method based on face 3D reconstruction and face alignment can adapt to various postures and expressions and effectively map the large posture and expression changes which can not be processed by traditional methods. This method can also use a general model to deal with all the faces and overcome the shortcomings of GAN method.
Description
technical field [0001] The invention relates to the field of image processing, in particular to a face image transformation method based on three-dimensional face reconstruction and face alignment. Background technique [0002] With the rise of the trend of artificial intelligence, especially the large-scale research and application of deep learning technology in the image field, the three-dimensional structure reconstruction and face alignment of faces have achieved better and better results. For example, the recently proposed pr_net has both accuracy and speed. Meet the practical requirements. [0003] In the field of entertainment and games, there are many needs to transform a face into another image, or to edit a face image, such as the synthesis of ancient style or Republic of China photos popular in WeChat Moments, synthesis of admission notices, etc.; In the post-processing of artistic photos, it is necessary to fine-tune the expression of the human face; in the prod...
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
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06K9/62
CPCG06V40/161G06V40/168G06F18/253
Inventor 周曦姚鹏飞
Owner 四川云从天府人工智能科技有限公司