CN107784678B - Cartoon face image generation method and device and terminal - Google Patents
Cartoon face image generation method and device and terminal Download PDFInfo
- Publication number
- CN107784678B CN107784678B CN201711093256.1A CN201711093256A CN107784678B CN 107784678 B CN107784678 B CN 107784678B CN 201711093256 A CN201711093256 A CN 201711093256A CN 107784678 B CN107784678 B CN 107784678B
- Authority
- CN
- China
- Prior art keywords
- cartoon
- noise
- face image
- attributes
- generating
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention provides a method, a device and a terminal for generating cartoon face images, wherein the method for generating the cartoon face images is executed based on a trained meta-network, and comprises the following steps: acquiring cartoon attributes appointed by a user; inputting the cartoon attribute into a preset meta-network to obtain a corresponding convolution kernel; generating noise information related to cartoon attributes; and obtaining the cartoon face image according to the convolution kernel convolution noise information. Compared with the prior art, according to the technical scheme provided by the invention, the complex image information can be integrated by utilizing the trained meta-network, the complete cartoon face image is quickly generated, the efficiency of drawing the cartoon face image is improved, the generation process is simple to operate, and the processing time is short. The cartoon face image obtained by the method has clear lines, bright colors and vivid effect, has no obvious difference with the cartoon image manually drawn by professionals, and meets the requirement of automatically drawing the cartoon image.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a method, a device and a terminal for generating a cartoon face image.
Background
In the modern society, the cartoon culture is used as an expression form of art, the relaxed and harmonious cultural atmosphere is gradually developed, and the thinking mode and the behavior of the teenagers in the modern times are greatly influenced. The cartoon image has the advantages of vivid and exaggerated color, vivid shape, high identification degree and affinity, can attract the attention of audiences, draw the distance from the audiences, relieve the pressure of people and meet the psychological needs of common people. Therefore, not only children like cartoon images, but also adults are willing to share and transmit the cartoon images. With the rapid development of digital media technology, the effect of drawing cartoon images through a computer has received much attention and research in many fields.
There are currently roughly two methods for using computers to render cartoon images. One method is to first do hand-drawing with a conventional tool, then scan to sufficient accuracy with a scanner, and then color and process in image processing software. Another method is to draw directly in computer software with a mouse or a tablet. The digital board is relatively flexible compared with a mouse, and parameter information such as main sensing technology, pressure sensing series, writing mode, personalized shape, recognition rate and the like is fed back to a computer through matching with a stylus, so that different pen touch effects are formed.
However, the drawing methods of these cartoon patterns are cumbersome, requiring a lot of time and effort to be invested by a professional trained drawing staff to manually make the cartoon patterns, and the generated images are greatly affected by personal skills and subjective consciousness, and are time-consuming and unable to fix a time-consuming period, which is not favorable for the development of media technology digitization.
Disclosure of Invention
In order to overcome the above technical problems or at least partially solve the above technical problems, the following technical solutions are proposed:
the invention provides a cartoon face image generation method, which comprises the following steps:
acquiring cartoon attributes appointed by a user;
inputting the cartoon attributes into a preset meta-network to obtain corresponding convolution kernels;
generating noise information related to the cartoon attributes;
and convolving the noise information according to the convolution kernel to obtain a cartoon face image.
Further, the step of obtaining the cartoon attribute specified by the user includes:
providing candidate items corresponding to various cartoon attributes;
and determining the cartoon attribute according to the detected option selected by the user.
Further, the step of generating the noise information related to the cartoon attribute comprises:
comparing the cartoon attribute with a plurality of preset noise elements to determine undefined noise elements in the cartoon attribute;
extracting relevant noise elements from a preset noise element database aiming at each undefined noise element;
and generating noise information related to the cartoon attributes according to the related noise elements.
Further, the step of convolving the noise information according to the convolution kernel to obtain a cartoon face image includes:
compressing the noise information;
obtaining a compressed cartoon face image according to the noise information after convolution and compression of the convolution kernel;
and decompressing the compressed cartoon face image.
Specifically, the cartoon attributes include at least one of: gender, organ size, organ color, facial expression, facial additional information.
In particular, the noise element comprises at least one of: imaging angle, character pose, organ shape, texture information.
The invention also provides a generating device of the cartoon face image, which comprises the following components:
the acquisition module is used for acquiring the cartoon attribute specified by the user;
the training module is used for inputting the cartoon attributes into a preset meta-network to obtain corresponding convolution kernels;
the generating module is used for generating noise information related to the cartoon attributes;
and the convolution module is used for convolving the noise information according to the convolution kernel to obtain the cartoon face image.
Further, the obtaining module is specifically configured to provide candidate items corresponding to multiple cartoon attributes; and determining the cartoon attribute according to the detected option selected by the user.
Further, the generation module is specifically configured to compare the cartoon attribute with a plurality of preset noise elements, and determine an undefined noise element in the cartoon attribute; extracting relevant noise elements from a preset noise element database aiming at each undefined noise element; and generating noise information related to the cartoon attributes according to the related noise elements.
Further, the convolution module is specifically configured to compress the noise information; obtaining a compressed cartoon face image according to the noise information after convolution and compression of the convolution kernel; and decompressing the compressed cartoon face image.
Specifically, the cartoon attributes include at least one of: gender, organ size, organ color, facial expression, facial additional information.
In particular, the noise element comprises at least one of: imaging angle, character pose, organ shape, texture information.
The invention also provides a terminal which comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein the processor executes the program to realize the generation method of the cartoon face image.
The invention also provides a readable storage medium, on which a computer program is stored, which when executed by a processor implements the method for generating the cartoon face image.
The invention provides a method, a device and a terminal for generating cartoon face images, wherein the method for generating the cartoon face images is executed based on a trained meta-network and comprises the following steps: acquiring cartoon attributes appointed by a user; inputting the cartoon attribute into a preset meta-network to obtain a corresponding convolution kernel; generating noise information related to cartoon attributes; and obtaining the cartoon face image according to the convolution kernel convolution noise information. Compared with the prior art, according to the technical scheme provided by the invention, the complex image information can be integrated by utilizing the trained meta-network, the complete cartoon face image is quickly generated, the efficiency of drawing the cartoon face image is improved, the generation process is simple to operate, and the processing time is short. The cartoon face image obtained by the method has clear lines, bright colors and vivid effect, has no obvious difference with the cartoon image manually drawn by professionals, and meets the requirement of automatically drawing the cartoon image.
Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
Fig. 1 is a schematic flow chart of a method for generating a cartoon face image according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of obtaining cartoon attributes according to an embodiment of the present invention;
fig. 3 is a schematic flow chart of generating noise information according to another embodiment of the present invention;
FIG. 4 is a flowchart illustrating a convolution method according to another embodiment of the present invention;
FIG. 5a is an exemplary diagram of a cartoon face image generated by an embodiment of the present invention;
FIG. 5b is an exemplary diagram of a cartoon face image drawn by hand in reality;
fig. 6 is a schematic frame diagram of a device for generating a cartoon face image according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present invention without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the invention provides a method for generating a cartoon face image, which comprises the following steps as shown in figure 1:
step S110: acquiring cartoon attributes appointed by a user;
step S120: inputting the cartoon attribute into a preset meta-network to obtain a corresponding convolution kernel;
step S130: generating noise information related to cartoon attributes;
step S140: and obtaining the cartoon face image according to the convolution kernel convolution noise information.
Therefore, the method for generating the cartoon face image provided by the embodiment of the invention can integrate complex image information by using the trained meta-network, quickly generate the complete cartoon face image, improve the efficiency of drawing the cartoon face image, and has the advantages of simple operation in the generation process and short processing time. The cartoon face image obtained by the method has clear lines, bright colors and vivid effect, has no obvious difference with the cartoon image manually drawn by professionals, and meets the requirement of automatically drawing the cartoon image.
In the embodiment of the invention, all the components of a cartoon face image are divided into two parts, namely cartoon attribute and noise information. The cartoon attribute refers to a part which is easy for a user to specify, and the component elements of the part generally have strong directivity and high identification degree, and specifically include but are not limited to:
gender, including male and female;
organ sizes, such as large eyes, small mouth, small nose, and the like;
organ colors, such as blue eye, brown skin, and the like;
facial expressions, such as laughing, smiling, crying, surprise, anger, shame, fear, and the like;
additional information on the face, such as a fiddle, a red hair, a pony tail, wearing glasses, etc.
The cartoon attributes can be set by those skilled in the art according to actual needs, and it should be understood that the cartoon attributes are only examples, and suitable changes based on these examples can also be applied to the present invention, and therefore, the present invention should also be included in the protection scope of the present invention.
In addition, the noise information refers to a part which is not easy to be specified by a user, and is matched with all cartoon attributes specified by the user when the cartoon face image is generated to form a complete cartoon face image. In other words, as long as the constituent elements required for generating the cartoon face image are not specified by the user, the noise information can be attributed to the constituent elements, and the noise information is generated in step S130, and the specific generation steps thereof will be described below.
Continuing with fig. 1, in step S110, the cartoon attributes specified by the user are obtained, where the cartoon attributes include all the cartoon attributes required by the user to draw a complete cartoon face image.
Optionally, as an embodiment of the present invention, as shown in fig. 2, the method includes:
step S111: and providing candidate items corresponding to various cartoon attributes.
And constructing a candidate list aiming at each cartoon attribute, and displaying the candidate list to the user in a candidate item mode for the user to select. For example, for the attribute "eye color", the "red", "yellow", "blue", "green", "black" waiting options may be provided for the user to select. Preferably, for the selection of the color class, an RGB color selector may be provided, visually giving the user the choice. Similarly, for candidate items with linearly changing size selection and the like, a progress bar dragging selection mode and the like can be provided. It can be understood that, in order to achieve more diversified imaging effects, the candidates corresponding to the cartoon attributes are also diversified as much as possible.
Specifically, all the cartoon attributes set according to actual needs construct corresponding candidate items, and a user can select a part of the cartoon attributes to specify, and can also select all the constructed cartoon attributes to specify. If some of the cartoon attributes are selected for designation, the non-designated cartoon attributes can be used as noise information.
Step S112: and determining the cartoon attribute according to the detected option selected by the user.
If the option selected by the user is detected, the cartoon attribute required to be formulated by the cartoon face image to be drawn by the user can be determined, so as to execute step S120.
Alternatively, if an empty item, that is, an option not selected by the user, is detected, recording is performed so as to execute step S130.
Alternatively, as another embodiment of the present invention, in step S110, the cartoon attribute specified by the user is obtained by recognizing the natural language input by the user. The method is not limited to the constructed candidate list and has richer experience.
Continuing with fig. 1, in step S120, the cartoon attributes are input into the preset meta-network to obtain the corresponding convolution kernels.
The predetermined meta-network is trained, for example, by a VGG-16 convolutional neural network (convolutional neural network).
Specifically, for any cartoon attribute, the sample images in the sample library are learned in advance. The number of sample images in the training library can be set by those skilled in the art according to actual needs, and is not limited herein. For the condition of combining a plurality of attributes, the meta-network can learn in sequence for each cartoon attribute, and can also directly learn different sample images with combined attributes.
In practical application, the training process of the meta-network is completed through multiple iterations until a predetermined convergence condition is met. The predetermined convergence condition can be set by those skilled in the art according to actual needs, and is not limited herein. For example, the predetermined convergence condition may include: the iteration times reach the preset iteration times; and/or the output value of the element network loss function is smaller than a preset threshold value; and/or the visual effect parameter of the image reaches the preset visual effect parameter. Specifically, whether the predetermined convergence condition is satisfied may be determined by determining whether the iteration number reaches a preset iteration number, whether the predetermined convergence condition is satisfied may be determined according to whether an output value of the meta-network loss function is smaller than a preset threshold, and whether the predetermined convergence condition is satisfied may be determined by determining whether a visual effect parameter of the image reaches a preset visual effect parameter.
Optionally, the one-iteration process comprises: randomly extracting a first sample image from the codebook, and generating a corresponding second sample image by using a convolution kernel corresponding to the first sample image; and obtaining a meta-network loss function according to the content loss between the second sample image and the first state sample image, and updating the weight parameter of the meta-network according to the meta-network loss function.
In a specific training process, a meta-network can be trained by using a stochastic gradient descent (stochastic gradient device) algorithm.
The cartoon attributes specified by the user and obtained in step S110 are input into the trained meta-network, so that the corresponding convolution kernel can be generated quickly.
As shown in fig. 1, in step S130, noise information related to the cartoon attribute is generated, and as can be seen from the above, the noise information is also a component of a cartoon face image, so the noise information must include a plurality of preset noise elements.
Specifically, noise elements include, but are not limited to: imaging angle, character pose, organ shape, texture information, color data, transparency data, smoothness data, reflectivity data, and the like.
Optionally, the noise element may include all cartoon attributes at the same time, and in the case that an empty item is detected, that is, all cartoon attributes are not specified by the user, the unspecified cartoon attributes may be synthesized as the noise element into the cartoon face image to ensure the integrity of the cartoon face image.
Optionally, as an embodiment of the present invention, acquiring only a cartoon attribute specified by a user, and not recording a detected empty item, as shown in fig. 3, includes:
step S131: and comparing the cartoon attribute with a plurality of preset noise elements to determine undefined noise elements in the cartoon attribute.
In order to determine the constituent elements required for the complete cartoon image, the noise elements that still need to be defined are determined by comparison, based on the cartoon attributes defined in step S110.
Step S132: extracting relevant noise elements from a preset noise element database aiming at each undefined noise element;
the preset noise element database stores a plurality of samples related to each preset noise element, wherein a person skilled in the art can set the number of samples and specific samples according to actual needs, which is not limited herein.
After the noise elements of the undefined noise elements are determined in step S131, for each undefined noise element, the relevant noise element is extracted from the preset noise element database in a manner of extracting according to a preset rule or randomly.
Step S133: and generating noise information related to the cartoon attributes according to the related noise elements.
It can be understood that the noise information generated by the noise element extracted in step S132 can be combined with the cartoon attribute specified by the user in step S130 to form a complete cartoon face image.
Continuing with fig. 1, in step S140, the cartoon face image is obtained according to the convolution kernel convolution noise information.
In practical application, the convolution kernel is a weight matrix, that is, the weight used in convolution is represented by a matrix, and the noise information is used as a large matrix, wherein each noise element corresponds to each element of the matrix, after the convolution kernel is obtained in step S120, the convolution kernel is used to perform convolution processing on the noise information, and the obtained cartoon face image is a complete cartoon face image.
Therefore, the cartoon face image obtained by the method is a complete image directly generated, and the meta-network obtained by training can be well suitable for noise information formed by various samples.
In order to achieve a smoother processing, as an embodiment of the present invention, as shown in fig. 4, optionally, the method includes:
step S141: the noise information is compressed.
Step S142: and obtaining a compressed cartoon face image according to the noise information after convolution and compression of the convolution kernel.
Step S143: and decompressing the compressed cartoon face image.
Because of the high complexity of noise information, providing realistic cartoon face images may require many high quality noise elements. Providing such noise information increases the load on the terminal convolution process, and therefore, compressing the noise information and then processing can reduce the requirements on the terminal hardware.
Moreover, the compressed noise information can keep the coherence characteristic of the noise elements, and reduce mutual interference, so that the generated cartoon face image is smoother.
The generation method of the cartoon face image provided by the embodiment of the invention is executed based on a trained meta-network, and comprises the following steps: acquiring cartoon attributes appointed by a user; inputting the cartoon attribute into a preset meta-network to obtain a corresponding convolution kernel; generating noise information related to cartoon attributes; and obtaining the cartoon face image according to the convolution kernel convolution noise information. Compared with the prior art, according to the technical scheme provided by the invention, the complex image information can be integrated by utilizing the trained meta-network, the complete cartoon face image is quickly generated, the efficiency of drawing the cartoon face image is improved, the generation process is simple to operate, and the processing time is short. In addition, fig. 5a and 5b respectively show an example of a cartoon face image generated by the embodiment of the invention and a cartoon face image drawn by hands in reality, so that the cartoon face image obtained by the method has clear lines, bright colors and vivid effect, has no obvious difference from the cartoon image drawn by a professional manually, and meets the requirement of automatically drawing the cartoon image.
An embodiment of the present invention further provides a generating apparatus of a cartoon face image, as shown in fig. 6, including:
the obtaining module 610 is configured to obtain a cartoon attribute specified by a user;
the training module 620 is configured to input the cartoon attribute into a preset meta-network to obtain a corresponding convolution kernel;
a generating module 630, configured to generate noise information related to the cartoon attribute;
and the convolution module 640 is used for convolving the noise information according to the convolution kernel to obtain the cartoon face image.
Optionally, the obtaining module 610 is specifically configured to provide candidate items corresponding to multiple cartoon attributes; and determining the cartoon attribute according to the detected option selected by the user.
Optionally, the generating module 630 is specifically configured to compare the cartoon attribute with a plurality of preset noise elements, and determine an undefined noise element in the cartoon attribute; extracting relevant noise elements from a preset noise element database aiming at each undefined noise element; and generating noise information related to the cartoon attributes according to the related noise elements.
Optionally, the convolution module 640 is specifically configured to compress the noise information; obtaining a compressed cartoon face image according to the noise information after convolution kernel convolution compression; and decompressing the compressed cartoon face image.
Specifically, the cartoon attributes include at least one of: gender, organ size, organ color, facial expression, facial additional information.
Specifically, the noise element includes at least one of: imaging angle, character pose, organ shape, texture information.
The embodiment of the invention also provides a terminal, which comprises a memory, a processor and a computer program which is stored on the memory and can be run on the processor, wherein the processor executes the program to realize the cartoon face image generation method of any one of the embodiments.
The embodiment of the invention also provides a readable storage medium, wherein a computer program is stored on the readable storage medium, and when the program is executed by a processor, the method for generating the cartoon face image is realized according to any one of the embodiments.
The generation device of the cartoon face image provided by the embodiment of the invention can integrate complex image information by utilizing the trained meta-network, quickly generate the complete cartoon face image, improve the efficiency of drawing the cartoon face image, and has simple operation in the generation process and short processing time. The cartoon face image obtained by the method has clear lines, bright colors and vivid effect, has no obvious difference with the cartoon image manually drawn by professionals, and meets the requirement of automatically drawing the cartoon image.
The generation device of the cartoon face image provided by the embodiment of the invention can be specific hardware on equipment or software or firmware installed on the equipment and the like. The device provided by the embodiment of the invention has the same realization principle and the same technical effect as the method embodiment. It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the foregoing systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
As used herein, the singular forms "a", "an", "the" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may also be present. Further, "connected" or "coupled" as used herein may include wirelessly connected or wirelessly coupled. As used herein, the term "and/or" includes all or any element and all combinations of one or more of the associated listed items.
It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
Those skilled in the art will appreciate that the present invention includes apparatus directed to performing one or more of the operations described in the present application. These devices may be specially designed and manufactured for the required purposes, or they may comprise known devices in general-purpose computers. These devices have stored therein computer programs that are selectively activated or reconfigured. Such a computer program may be stored in a device (e.g., computer) readable medium, including, but not limited to, any type of disk including floppy disks, hard disks, optical disks, CD-ROMs, and magnetic-optical disks, ROMs (Read-Only memories), RAMs (Random Access memories), EPROMs (Erasable Programmable Read-Only memories), EEPROMs (Electrically Erasable Programmable Read-Only memories), flash memories, magnetic cards, or optical cards, or any type of media suitable for storing electronic instructions, and each coupled to a bus. That is, a readable medium includes any medium that stores or transmits information in a form readable by a device (e.g., a computer).
It will be understood by those within the art that each block of the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations, can be implemented by computer program instructions. Those skilled in the art will appreciate that the computer program instructions may be implemented by a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, implement the features specified in the block or blocks of the block diagrams and/or flowchart illustrations of the present disclosure.
Those of skill in the art will appreciate that various operations, methods, steps in the processes, acts, or solutions discussed in the present application may be alternated, modified, combined, or deleted. Further, various operations, methods, steps in the flows, which have been discussed in the present application, may be interchanged, modified, rearranged, decomposed, combined, or eliminated. Further, steps, measures, schemes in the various operations, methods, procedures disclosed in the prior art and the present invention can also be alternated, changed, rearranged, decomposed, combined, or deleted.
The foregoing is only a partial embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.
Claims (12)
1. A method for generating cartoon face image, the method comprises:
acquiring cartoon attributes appointed by a user;
inputting the cartoon attributes into a preset meta-network to obtain corresponding convolution kernels;
generating noise information related to the cartoon attributes;
convolving the noise information according to the convolution kernel to obtain a cartoon face image;
wherein the step of generating the noise information related to the cartoon attributes comprises:
comparing the cartoon attribute with a plurality of preset noise elements to determine undefined noise elements in the cartoon attribute;
extracting relevant noise elements from a preset noise element database aiming at each undefined noise element;
and generating noise information related to the cartoon attributes according to the related noise elements.
2. The method of claim 1, wherein the step of obtaining the user-specified cartoon attributes comprises:
providing candidate items corresponding to various cartoon attributes;
and determining the cartoon attribute according to the detected option selected by the user.
3. The method as claimed in claim 1, wherein the step of convolving the noise information with the convolution kernel to obtain a cartoon face image comprises:
compressing the noise information;
obtaining a compressed cartoon face image according to the noise information after convolution and compression of the convolution kernel;
and decompressing the compressed cartoon face image.
4. The generation method according to claim 1, characterized in that the cartoon attributes comprise at least one of the following: gender, organ size, organ color, facial expression, facial additional information.
5. The generation method according to claim 1, characterized in that the noise element comprises at least one of: imaging angle, character pose, organ shape, texture information.
6. An apparatus for generating a cartoon face image, comprising:
the acquisition module is used for acquiring the cartoon attribute specified by the user;
the training module is used for inputting the cartoon attributes into a preset meta-network to obtain corresponding convolution kernels;
the generating module is used for generating noise information related to the cartoon attributes;
the convolution module is used for convolving the noise information according to the convolution kernel to obtain a cartoon face image;
the generation module is specifically configured to compare the cartoon attribute with a plurality of preset noise elements, and determine an undefined noise element in the cartoon attribute; extracting relevant noise elements from a preset noise element database aiming at each undefined noise element; and generating noise information related to the cartoon attributes according to the related noise elements.
7. The generation apparatus as claimed in claim 6, wherein the obtaining module is specifically configured to provide candidate items corresponding to a plurality of cartoon attributes; and determining the cartoon attribute according to the detected option selected by the user.
8. The generation apparatus according to claim 6, wherein the convolution module is specifically configured to compress the noise information; obtaining a compressed cartoon face image according to the noise information after convolution and compression of the convolution kernel; and decompressing the compressed cartoon face image.
9. The generation apparatus according to claim 6, wherein the cartoon attributes comprise at least one of: gender, organ size, organ color, facial expression, facial additional information.
10. The generation apparatus of claim 6, wherein the noise element comprises at least one of: imaging angle, character pose, organ shape, texture information.
11. A terminal comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method for generating a cartoon face image according to any one of claims 1 to 5 when executing the program.
12. A readable storage medium on which a computer program is stored, which program, when being executed by a processor, is adapted to carry out a method of generating a cartoon face image according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711093256.1A CN107784678B (en) | 2017-11-08 | 2017-11-08 | Cartoon face image generation method and device and terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711093256.1A CN107784678B (en) | 2017-11-08 | 2017-11-08 | Cartoon face image generation method and device and terminal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107784678A CN107784678A (en) | 2018-03-09 |
| CN107784678B true CN107784678B (en) | 2021-11-12 |
Family
ID=61432485
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711093256.1A Active CN107784678B (en) | 2017-11-08 | 2017-11-08 | Cartoon face image generation method and device and terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107784678B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110276809A (en) * | 2018-03-15 | 2019-09-24 | 深圳市紫石文化传播有限公司 | Method and apparatus for face image processing |
| CN108596839A (en) * | 2018-03-22 | 2018-09-28 | 中山大学 | A kind of human-face cartoon generation method and its device based on deep learning |
| CN108765516A (en) * | 2018-03-26 | 2018-11-06 | 阳光暖果（北京）科技发展有限公司 | A kind of face region colouring method based on thumbnail |
| CN109919018A (en) * | 2019-01-28 | 2019-06-21 | 浙江英索人工智能科技有限公司 | Image eyes based on reference picture automatically open method and device |
| CN109978759B (en) * | 2019-03-27 | 2023-01-31 | 北京市商汤科技开发有限公司 | Image processing method and device and training method and device of image generation network |
| CN112883759B (en) * | 2019-11-29 | 2023-09-26 | 杭州海康威视数字技术股份有限公司 | Method for detecting image noise of biological feature part |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101527049A (en) * | 2009-03-31 | 2009-09-09 | 西安交通大学 | Generating method of multiple-style face cartoon based on sample learning |
| CN102663806A (en) * | 2012-03-02 | 2012-09-12 | 西安交通大学 | Artistic-vision-based cartoon stylized rendering method of image |
| JP2013003817A (en) * | 2011-06-16 | 2013-01-07 | V-Sync Co Ltd | Environment understanding type control system by face recognition |
| CN104112113A (en) * | 2013-04-19 | 2014-10-22 | 无锡南理工科技发展有限公司 | Improved characteristic convolutional neural network image identification method |
| CN104240274A (en) * | 2014-09-29 | 2014-12-24 | 小米科技有限责任公司 | Face image processing method and device |
| CN106599832A (en) * | 2016-12-09 | 2017-04-26 | 重庆邮电大学 | Method for detecting and recognizing various types of obstacles based on convolution neural network |
| CN106599837A (en) * | 2016-12-13 | 2017-04-26 | 北京智慧眼科技股份有限公司 | Face identification method and device based on multi-image input |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106295476B (en) * | 2015-05-29 | 2019-05-17 | 腾讯科技（深圳）有限公司 | Face key point localization method and device |
-
2017
- 2017-11-08 CN CN201711093256.1A patent/CN107784678B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101527049A (en) * | 2009-03-31 | 2009-09-09 | 西安交通大学 | Generating method of multiple-style face cartoon based on sample learning |
| JP2013003817A (en) * | 2011-06-16 | 2013-01-07 | V-Sync Co Ltd | Environment understanding type control system by face recognition |
| CN102663806A (en) * | 2012-03-02 | 2012-09-12 | 西安交通大学 | Artistic-vision-based cartoon stylized rendering method of image |
| CN104112113A (en) * | 2013-04-19 | 2014-10-22 | 无锡南理工科技发展有限公司 | Improved characteristic convolutional neural network image identification method |
| CN104240274A (en) * | 2014-09-29 | 2014-12-24 | 小米科技有限责任公司 | Face image processing method and device |
| CN106599832A (en) * | 2016-12-09 | 2017-04-26 | 重庆邮电大学 | Method for detecting and recognizing various types of obstacles based on convolution neural network |
| CN106599837A (en) * | 2016-12-13 | 2017-04-26 | 北京智慧眼科技股份有限公司 | Face identification method and device based on multi-image input |
Non-Patent Citations (1)
| Title |
|---|
| 《Convolutional Network for Attribute-driven and Identity-preserving Human Face Generation》;Mu Li等;《arXiv》;20160423;1-9 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107784678A (en) | 2018-03-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107784678B (en) | Cartoon face image generation method and device and terminal | |
| US10489683B1 (en) | Methods and systems for automatic generation of massive training data sets from 3D models for training deep learning networks | |
| US20180374242A1 (en) | Avatar digitization from a single image for real-time rendering | |
| EP3885965B1 (en) | Image recognition method based on micro facial expressions, apparatus and related device | |
| WO2020258668A1 (en) | Facial image generation method and apparatus based on adversarial network model, and nonvolatile readable storage medium and computer device | |
| CN107784630B (en) | Method, device and terminal for turning attributes of face image | |
| CN106326857A (en) | Gender identification method and gender identification device based on face image | |
| CN106295591A (en) | Gender identification method based on facial image and device | |
| CN109002763B (en) | Method and device for simulating human face aging based on homologous continuity | |
| US20220164852A1 (en) | Digital Imaging and Learning Systems and Methods for Analyzing Pixel Data of an Image of a Hair Region of a User's Head to Generate One or More User-Specific Recommendations | |
| US11507781B2 (en) | Methods and systems for automatic generation of massive training data sets from 3D models for training deep learning networks | |
| KR20190076858A (en) | Improved age modelling method | |
| CN112489129A (en) | Pose recognition model training method and device, pose recognition method and terminal equipment | |
| CN109241869A (en) | The recognition methods of answering card score, device and terminal device | |
| CN111325173A (en) | Hair type identification method and device, electronic equipment and storage medium | |
| CN110019662A (en) | A kind of label method for reconstructing and device | |
| CN116469147A (en) | Facial expression migration method and device, electronic equipment and storage medium | |
| CN114333018A (en) | Shaping information recommendation method and device and electronic equipment | |
| CN117011449A (en) | Reconstruction method and device of three-dimensional face model, storage medium and electronic equipment | |
| CN116452601A (en) | Virtual fitting method, virtual fitting device, electronic equipment and storage medium | |
| CN114494529A (en) | Face attribute editing method based on cost sensitive learning | |
| JP6320844B2 (en) | Apparatus, program, and method for estimating emotion based on degree of influence of parts | |
| CN108229491A (en) | The method, apparatus and equipment of detection object relationship from picture | |
| CN107958229B (en) | Face recognition method, device and equipment based on neighbor keeping low-rank representation | |
| WO2021031242A1 (en) | Character verification method and apparatus, computer device, and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |