Unlock instant, AI-driven research and patent intelligence for your innovation.

Three-dimensional cartoon model automatic generation method and system based on single face image

A face image, automatic generation technology, applied in 3D image processing, image data processing, neural learning methods, etc., can solve the problems of automatic generation, such as less work, cumbersome production, time-consuming, etc., to solve the problem of non-leadability , rich texture style effect

Pending Publication Date: 2022-01-28
BEIHANG UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

3D cartoons are suitable for these applications, but they can only be created by artists with 3D modeling skills, and the production is tedious and time-consuming. Therefore, automatic generation of 3D exaggerated style human faces is a meaningful but lacking research direction
Generating 3D caricatures from 2D caricatures or ordinary photos is similar to face reconstruction, but few works address the automatic generation of 3D caricatures from face pictures

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
  • Three-dimensional cartoon model automatic generation method and system based on single face image
  • Three-dimensional cartoon model automatic generation method and system based on single face image
  • Three-dimensional cartoon model automatic generation method and system based on single face image

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0022] Such as figure 1 As shown, a kind of method for automatically generating a three-dimensional caricature model based on a single face image provided by the embodiment of the present invention comprises the following steps:

[0023] Step S1: Based on the face image of the same identity, the corresponding caricature image and textured 3D caricature model are constructed to form a caricature dataset. The face image and the caricature image have the same facial pose;

[0024] Step S2: Input the textured 3D comic model into the graph convolutional autoencoder, which includes an encoder and decoder with grid convolution operation and grid sampling operation, and uses the graph convolutional autoencoder to The textured 3D comic model is encoded to obtain the geometric and texture representation of the 3D comic model;

[0025] Step S3: Input the face image into the ResNet network to obtain the vector of the facial pose vector and the geometric and texture representation of the ...

Embodiment 2

[0076] Such as Figure 6 As shown, the embodiment of the present invention provides a kind of automatic generation system of three-dimensional caricature model based on single face image, comprises following module:

[0077]Construct comic data set module 51, be used for constructing corresponding comic image and textured three-dimensional comic model based on the face image of the same identity, form comic data set, described human face image and comic image have identical facial pose;

[0078] Obtain the geometric and texture representation module 52 of the three-dimensional comic model, for inputting the textured three-dimensional comic model into an autoencoder based on graph convolution, the self-encoder includes a grid convolution operation and a grid sampling operation An encoder and a decoder, using the graph convolutional self-encoder to encode the textured three-dimensional caricature model to obtain a geometric and texture representation of the three-dimensional car...

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 invention relates to a three-dimensional cartoon model automatic generation method and system based on a single face image, and the method comprises the steps: S1, constructing a corresponding cartoon image and a texture-containing three-dimensional cartoon model based on face images of the same identity, and forming a cartoon data set; S2, inputting the texture-containing three-dimensional cartoon model into an auto-encoder based on graph convolution to obtain geometric and texture representations of the three-dimensional cartoon model; S3, inputting the face image into a ResNet network to obtain a face posture vector and a vector expressed by geometry and texture of the three-dimensional cartoon model; S4, obtaining a generated three-dimensional cartoon model from the vectors represented by the geometry and the texture of the three-dimensional cartoon model by using a decoder, converting the generated three-dimensional cartoon model into a face posture which is the same as that of the face image by using the face posture vector, and rendering the generated three-dimensional cartoon model by using a differentiatable renderer. According to the method provided by the invention, the three-dimensional cartoon model with exaggerated geometry and rich texture styles can be efficiently and quickly generated.

Description

technical field [0001] The invention relates to the fields of computer graphics and image processing, in particular to a method and system for automatically generating a three-dimensional caricature model based on a single face image. Background technique [0002] A caricature is a rendered image that expresses the most distinctive features of a person through exaggeration, simplification, and abstraction. They are also used to express irony and humor about political and social issues. The caricature drawn by the artist is a 2D image. Although widely used, they are insufficient for many applications, such as animation, virtual reality, and 3D printing, for which 3D information is essential. 3D cartoons are suitable for these applications, but they can only be created by artists with 3D modeling skills, and the production is tedious and time-consuming. Therefore, automatic generation of 3D exaggerated-style human faces is a meaningful but lacking research direction. Genera...

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
IPC IPC(8): G06T17/00G06T15/00G06N3/04G06N3/08
CPCG06T17/00G06T15/005G06N3/08G06N3/045
Inventor 潘俊君黄美佳
Owner BEIHANG UNIV