CN111311707B - Painting method and device - Google Patents
Painting method and device Download PDFInfo
- Publication number
- CN111311707B CN111311707B CN202010147722.5A CN202010147722A CN111311707B CN 111311707 B CN111311707 B CN 111311707B CN 202010147722 A CN202010147722 A CN 202010147722A CN 111311707 B CN111311707 B CN 111311707B
- Authority
- CN
- China
- Prior art keywords
- state information
- manuscript
- preset
- user
- complete
- 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
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B44—DECORATIVE ARTS
- B44D—PAINTING OR ARTISTIC DRAWING, NOT OTHERWISE PROVIDED FOR; PRESERVING PAINTINGS; SURFACE TREATMENT TO OBTAIN SPECIAL ARTISTIC SURFACE EFFECTS OR FINISHES
- B44D3/00—Accessories or implements for use in connection with painting or artistic drawing, not otherwise provided for; Methods or devices for colour determination, selection, or synthesis, e.g. use of colour tables
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B44—DECORATIVE ARTS
- B44D—PAINTING OR ARTISTIC DRAWING, NOT OTHERWISE PROVIDED FOR; PRESERVING PAINTINGS; SURFACE TREATMENT TO OBTAIN SPECIAL ARTISTIC SURFACE EFFECTS OR FINISHES
- B44D3/00—Accessories or implements for use in connection with painting or artistic drawing, not otherwise provided for; Methods or devices for colour determination, selection, or synthesis, e.g. use of colour tables
- B44D3/22—Implements or apparatus for special techniques, e.g. for painting lines, for pouring varnish; Batik pencils
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
- Facsimiles In General (AREA)
Abstract
The invention discloses a drawing method and a drawing device, comprising the following steps: training a preset network model by using a preset manuscript; receiving a part of the draft of the current picture drawn by the user, and updating the part of the draft by using a trained preset network model; displaying the updated complete manuscript to confirm whether the user is satisfied with the complete manuscript, wherein the updated complete manuscript comprises an updated partial manuscript; and after the user is satisfied with the complete manuscript, outputting the complete manuscript. The method comprises the steps that whether a user is satisfied or not is determined by displaying the user in the process of updating part of the manuscript by the preset network model, if the user is satisfied, the updated complete manuscript is output, and if the user is not satisfied, the preset network model is continuously updated until the user is satisfied, so that the user can finally obtain the ideal manuscript according to the self requirement and preference, and the problem that the user cannot generate the ideal manuscript of the user because the user cannot participate in the process of updating the manuscript by the model in the prior art is solved.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a painting method and device.
Background
Along with the development of science and technology, modern people are more and more in the field of computers, and deep learning brings new rounds of artificial intelligence surge, and is widely focused in a plurality of fields. Especially in the field of graphic images, applications such as face recognition and autopilot are gradually entering our lives. Deep learning is a method for performing characterization learning on data in machine learning. The mechanisms mimicking the human brain interpret data such as images, sounds, text, etc. The advantage is that the feature is obtained manually by using an unsupervised or semi-supervised feature learning and hierarchical feature extraction efficient algorithm instead. In the painting field, a user only needs to input an original image and a final image into a training model for training, and then input a blank painting into a network model to obtain a corresponding complete painting, but the technology has the following defects: because the network model is trained in advance, the blank picture is input into the network model to output a complete draft once, and in the process, the user does not participate in the process, so that an output result and an ideal result of the user come in and go out, and the use experience of the user is reduced.
Disclosure of Invention
Aiming at the displayed problems, the method is based on training a preset network model by using a preset manuscript, then inputting part of the current picture of the user into the preset network model for updating, and outputting the final complete picture after confirming the satisfaction of the user in the updating process.
A method of drawing comprising the steps of:
training a preset network model by using a preset manuscript;
receiving a partial draft of a current picture drawn by a user, and updating the partial draft by using a trained preset network model;
displaying the updated complete manuscript to confirm whether the user is satisfied with the complete manuscript, wherein the updated complete manuscript comprises the updated partial manuscript;
and after the user is satisfied with the complete manuscript, outputting the complete manuscript.
Preferably, the training the preset network model by using the preset manuscript includes:
acquiring a preset number of preset drawings;
acquiring N images of each preset drawing at intervals of preset time intervals or preset steps in the process from the beginning of drawing to the end of drawing;
determining first preset state information, second preset state information and/or third preset state information of the N images from the first image to the last image;
taking the first preset state information, the second preset state information and the first, second and third, fourth and fifth preset state information as input of the preset network model, and receiving first output state information, second output state information and third, fourth and fifth output state information output by the preset network model;
comparing the second preset state information with the first output state information to obtain first similarity, comparing the third preset state information with the second output state information to obtain second similarity, and repeating the steps to obtain N-1 similarities;
and training the preset network model by taking the N-1 similarities as a loss function.
Preferably, the receiving a part of the draft of the current picture drawn by the user and updating the part of the draft by using the trained preset network model includes:
determining first current state information of the partial drawing, wherein the first current state information is all information in a canvas area on a current drawing board;
inputting the first current state information into the trained preset network model to obtain second current output state information corresponding to the first current state information;
and updating the partial manuscript for the first time according to the second current output state information.
Preferably, the method further comprises:
when the first updating of the partial manuscript is finished by utilizing the second current output state information, displaying the updated partial manuscript so that the user modifies the updated partial manuscript;
if the modified updated partial manuscript is received, determining third current state information of the updated partial manuscript, and taking the third current state information as input of a trained preset network model to obtain fourth current output state information corresponding to the third current state information;
performing a second update on the updated partial manuscript after modification according to the fourth current output state information;
repeating the steps to automatically update until the user is satisfied;
and if the modified updated partial manuscript is not received, taking the second current output state information as the input of the trained preset network model, and continuing updating until the user is satisfied.
Preferably, the displaying the updated complete draft of the current picture to confirm whether the user is satisfied with the complete draft includes:
confirming whether the complete manuscript is a final manuscript;
if yes, optimizing the complete sketch and displaying the complete sketch, and acquiring satisfaction of the user;
otherwise, continuing to update the complete manuscript until the complete manuscript is determined to be the final manuscript.
A painting apparatus, the apparatus comprising:
the training module is used for training the preset network model by utilizing the preset drawing manuscript;
the first updating module is used for receiving partial manuscripts of the current picture drawn by the user and updating the partial manuscripts by utilizing the trained preset network model;
the confirmation module is used for displaying the updated complete manuscript so as to confirm whether the user is satisfied with the complete manuscript, wherein the updated complete manuscript comprises the updated partial manuscript;
and the output module is used for outputting the complete manuscript after determining that the user is satisfied with the complete manuscript.
Preferably, the training module includes:
the first acquisition submodule is used for acquiring a preset number of preset drawings;
the second acquisition submodule is used for acquiring N images of each preset drawing at intervals of preset time or at preset steps in the process from the beginning of drawing to the end of drawing;
a first determining sub-module for determining first preset state information, second preset state information, and third preset state information of the N images from a first image to a last image;
the receiving sub-module is used for taking the first preset state information, the second preset state information and the third preset state information as input of the preset network model and receiving the first output state information, the second output state information and the third preset state information, wherein the first output state information, the second output state information and the third output state information are output by the preset network model;
the comparison sub-module is used for comparing the second preset state information with the first output state information to obtain first similarity, comparing the third preset state information with the second output state information to obtain second similarity, and repeating the steps to obtain N-1 similarities;
and the training sub-module is used for training the preset network model by taking the N-1 similarities as a loss function.
Preferably, the first updating module includes:
the second determining submodule is used for determining first current state information of the partial drawing, wherein the first current state information is all information in a canvas area on a current drawing board;
a fourth obtaining sub-module, configured to input the first current state information into the trained preset network model to obtain second current output state information corresponding to the first current state information;
and the first updating sub-module is used for updating the partial manuscript for the first time according to the second output current state information.
Preferably, the apparatus further comprises:
the display module is used for displaying the updated partial manuscript so that the user modifies the updated partial manuscript when the first updating of the partial manuscript is finished by utilizing the second current output state information;
the determining module is used for determining third current state information of the updated partial manuscript if the updated partial manuscript is received, and taking the third current state information as input of a trained preset network model to obtain fourth current output state information corresponding to the third current state information;
the second updating module is used for updating the updated part of the manuscript after modification for the second time according to the fourth current output state information;
repeating the steps to automatically update until the user is satisfied;
and the third updating module is used for taking the second current output state information as the input of the trained preset network model and continuing updating until the user is satisfied if the modified updated partial draft is not received.
Preferably, the confirmation module includes:
a confirmation sub-module for confirming whether the complete manuscript is a final manuscript;
the processing sub-module is used for carrying out optimization processing and displaying on the complete manuscript to obtain the satisfaction degree of the user if the complete manuscript is confirmed to be the final manuscript by the confirmation sub-module;
and the second updating sub-module is used for continuously updating the complete manuscript until the complete manuscript is determined to be the final manuscript if the confirmation sub-module confirms that the complete manuscript is not the final manuscript.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention may be realized and attained by the structure particularly pointed out in the written description and drawings.
The technical scheme of the invention is further described in detail through the drawings and the embodiments.
Drawings
FIG. 1 is a workflow diagram of a painting method provided by the present invention;
FIG. 2 is another workflow diagram of a painting method provided by the present invention;
FIG. 3 is a block diagram of a drawing device according to the present invention;
fig. 4 is another structural diagram of a drawing device provided by the invention.
Detailed Description
Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples are not representative of all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with some aspects of the present disclosure as detailed in the accompanying claims.
Along with the development of science and technology, modern people are more and more in the field of computers, and deep learning brings new rounds of artificial intelligence surge, and is widely focused in a plurality of fields. Especially in the field of graphic images, applications such as face recognition and autopilot are gradually entering our lives. Deep learning is a method for performing characterization learning on data in machine learning. The mechanisms mimicking the human brain interpret data such as images, sounds, text, etc. The advantage is that the feature is obtained manually by using an unsupervised or semi-supervised feature learning and hierarchical feature extraction efficient algorithm instead. In the painting field, a user only needs to input an original image and a final image into a training model for training, and then input a blank painting into a network model to obtain a corresponding complete painting, but the technology has the following defects: because the network model is trained in advance, the blank picture is input into the network model to output a complete draft once, and in the process, the user does not participate in the process, so that an output result and an ideal result of the user come in and go out, and the use experience of the user is reduced. In order to solve the above-mentioned problems, the present embodiment discloses a method for training a preset network model based on using a preset draft, then inputting a part of the draft of the current picture of the user into the preset network model for updating at the heart, and outputting the final complete draft after confirming the satisfaction of the user in the updating process.
A drawing method, as shown in fig. 1, comprising the steps of:
step S101, training a preset network model by using a preset draft;
step S102, receiving a part of a draft of a current picture drawn by a user, and updating the part of the draft by using a trained preset network model;
step S103, displaying the updated complete manuscript to confirm whether the user is satisfied with the complete manuscript, wherein the updated complete manuscript comprises an updated partial manuscript;
and step S104, after the user is satisfied with the complete draft, outputting the complete draft.
The working principle of the technical scheme is as follows: training a preset network model by using a preset manuscript, then receiving and receiving a part of the manuscript of the current picture drawn by the user, updating the part of the manuscript by using the trained preset network model, displaying the updated complete manuscript after updating so as to confirm whether the user is satisfied with the complete manuscript, and outputting the complete manuscript after confirming that the user is satisfied with the complete manuscript.
The beneficial effects of the technical scheme are as follows: the method has the advantages that whether the user is satisfied is determined by displaying the user in the process of updating part of the manuscript by the preset network model, if the user is satisfied, the updated complete manuscript is output, and if the user is not satisfied, the preset network model is continuously updated until the user is satisfied, so that the user can finally obtain the ideal manuscript according to the self requirements and the favorites, the problem that the user cannot generate the ideal manuscript of the user because the user cannot participate in the process of updating the manuscript by the model in the prior art is solved, and the use experience of the user is improved.
In one embodiment, training the preset network model with the preset manuscript includes:
acquiring a preset number of preset drawings;
acquiring N images of each preset drawing at intervals of preset time intervals or preset steps in the process from the beginning of drawing to the end of drawing;
determining first preset state information, second preset state information, and third preset state information of N images from a first image to a last image;
taking the first preset state information, the second preset state information and the third and fourth preset state information as input of a preset network model, and receiving the first output state information, the second output state information and the third and fourth output state information output by the preset network model;
comparing the second preset state information with the first output state information to obtain a first similarity, comparing the third preset state information with the second output state information to obtain a second similarity, and repeating the steps to obtain N-1 similarities;
and training the preset network model by taking the N-1 similarities as a loss function.
In this embodiment, the preset time interval may be 30s, that is, the drawing image content of the preset drawing is obtained every 30s from the beginning of drawing to the end of drawing, and the first state information is the completion status of each drawing image content, for example: the first painting image content completes one sixth of the painting end image, and the second painting image content completes one fifth of the painting end image, and the similarity refers to the similarity between original state information of the picture and output state information input into a preset network model and output, and the similarity is used as a loss function to enable the preset network model to converge.
The beneficial effects of the technical scheme are as follows: the preset network model is trained by using the preset state information and the similarity value of the output state information, so that when the preset network model receives the current partial drawing of a user, the current state information of the partial drawing is directly determined, further, the image content of the next drawing can be determined according to the current state information and the similarity difference value, the current partial drawing is automatically updated, the preset network model is trained by using N images of the preset drawing at intervals of preset time intervals or preset steps in the process from the beginning of drawing to the end of drawing, and the generation process can be participated by the user.
In one embodiment, as shown in fig. 2, receiving a partial draft of a current picture drawn by a user, and updating the partial draft by using a trained preset network model includes:
step S201, determining first current state information of a part of drawing, wherein the first current state information is all information in a canvas area on a current drawing board;
step S202, inputting the first current state information into a trained preset network model to obtain second current output state information corresponding to the first current state information;
step S203, the partial manuscript is updated for the first time according to the second current output state information.
The beneficial effects of the technical scheme are as follows: the second current output state information corresponding to the first current state information can be accurately determined, and after the second current output state information is determined, part of the manuscript can be updated directly according to the second current output state information, so that the updating speed and the updating efficiency are improved.
In one embodiment, the method further comprises:
when the first updating of the partial manuscript is finished by utilizing the second current output information, displaying the updated partial manuscript so as to enable a user to modify the updated partial manuscript;
if the modified updated partial manuscript is received, determining third current state information of the updated partial manuscript, and taking the third current state information as input of a trained preset network model to obtain fourth current output state information corresponding to the third current state information;
carrying out second updating on the updated part of the manuscript after modification according to the fourth current output state information;
repeating the steps to automatically update until the user is satisfied;
if the modified updated partial draft is not received, the second current output state information is used as the input of the trained preset network model, and the updating is continued until the user is satisfied.
The beneficial effects of the technical scheme are as follows: the user can modify the updated partial manuscript according to the preference and the requirement, and the modified partial manuscript is continuously updated by the preset network model after the modification until the user is satisfied, so that the user can be integrated into the manuscript updating process, and the use experience of the user is further improved.
In one embodiment, displaying the updated complete drawing of the current drawing to confirm whether the user is satisfied with the complete drawing includes:
confirming whether the complete manuscript is a final manuscript;
if yes, optimizing and displaying the complete sketch, and acquiring satisfaction degree of the user;
otherwise, continuing to update the complete manuscript until the complete manuscript is determined to be the final manuscript.
The beneficial effects of the technical scheme are as follows: the complete manuscript is more perfect, and the ideal manuscript of the user is finally obtained according to the needs of the user.
A painting apparatus, as shown in fig. 3, comprising:
the training module 301 is configured to train the preset network model by using a preset drawing;
a first updating module 302, configured to receive a part of a drawing of a current drawing drawn by a user, and update the part of the drawing by using a trained preset network model;
a confirmation module 303, configured to display the updated complete manuscript to confirm whether the user is satisfied with the complete manuscript, where the updated complete manuscript includes an updated partial manuscript;
and the output module 304 is configured to output the complete drawing after determining that the user is satisfied with the complete drawing.
In one embodiment, a training module includes:
the first acquisition submodule is used for acquiring a preset number of preset drawings;
the second acquisition submodule is used for acquiring N images of each preset drawing at intervals of preset time or at preset steps in the process from the beginning of drawing to the end of drawing;
a first determining sub-module for determining first preset state information, second preset state information, and third preset state information of N images from a first image to a last image;
the receiving sub-module is used for taking the first preset state information, the second preset state information and the third preset state information as input of a preset network model and receiving the first output state information, the second output state information and the third output state information output by the preset network model;
the comparison sub-module is used for comparing the second preset state information with the first output state information to obtain first similarity, comparing the third preset state information with the second output state information to obtain second similarity, and repeating the steps to obtain N-1 similarities;
and the training sub-module is used for training the preset network model by taking the N-1 similarities as a loss function.
In one embodiment, as shown in fig. 4, the first update module includes:
a second determining submodule 3021, configured to determine first current status information of a part of the drawing, where the first current status information is all information in a canvas area on a current drawing board;
a fourth obtaining submodule 3022, configured to input the first current state information into the trained preset network model to obtain second current output state information corresponding to the first current state information;
and a first updating sub-module 3023, configured to update the partial manuscript for the first time according to the second current output status information.
In one embodiment, the apparatus further comprises:
the display module is used for displaying the updated partial manuscript so as to enable a user to modify the updated partial manuscript when the first updating of the partial manuscript is finished by utilizing the second current output state information;
the determining module is used for determining third current state information of the updated partial manuscript if the updated partial manuscript is received, and taking the third current state information as input of a trained preset network model to obtain fourth current output state information corresponding to the third current state information;
the second updating module is used for carrying out second updating on the modified updated partial manuscript according to the fourth current output state information;
repeating the steps to automatically update until the user is satisfied;
and the third updating module is used for taking the second current output state information as the input of the trained preset network model and continuing updating until the user is satisfied if the modified updated partial draft is not received.
In one embodiment, the validation module includes:
the confirming sub-module is used for confirming whether the complete drawing is a final drawing;
the processing sub-module is used for optimizing and displaying the complete manuscript to obtain the satisfaction degree of the user if the confirmation sub-module confirms that the complete manuscript is the final manuscript;
and the second updating sub-module is used for continuously updating the complete manuscript until the complete manuscript is determined to be the final manuscript if the confirmation sub-module confirms that the complete manuscript is not the final manuscript.
It will be appreciated by those skilled in the art that the first and second aspects of the present invention refer to different phases of application.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This application is intended to cover any adaptations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
Claims (6)
1. A method of drawing, comprising the steps of:
training a preset network model by using a preset manuscript;
receiving a partial draft of a current picture drawn by a user, and updating the partial draft by using a trained preset network model;
displaying the updated complete manuscript to confirm whether the user is satisfied with the complete manuscript, wherein the updated complete manuscript comprises the updated partial manuscript;
after determining that the user is satisfied with the complete drawing, outputting the complete drawing;
the training of the preset network model by using the preset manuscript comprises the following steps:
acquiring a preset number of preset drawings;
acquiring N images of each preset drawing at intervals of preset time intervals or preset steps in the process from the beginning of drawing to the end of drawing;
determining first preset state information, second preset state information and/or third preset state information of the N images from the first image to the last image;
taking the first preset state information, the second preset state information and the first, second and third, fourth and fifth preset state information as input of the preset network model, and receiving first output state information, second output state information and third, fourth and fifth output state information output by the preset network model;
comparing the second preset state information with the first output state information to obtain first similarity, comparing the third preset state information with the second output state information to obtain second similarity, and repeating the steps to obtain N-1 similarities;
training the preset network model by taking the N-1 similarities as a loss function;
the step of receiving the partial manuscript of the current picture drawn by the user and updating the partial manuscript by using the trained preset network model comprises the following steps:
determining first current state information of the partial drawing, wherein the first current state information is all information in a canvas area on a current drawing board;
inputting the first current state information into the trained preset network model to obtain second current output state information corresponding to the first current state information;
and updating the partial manuscript for the first time according to the second current output state information.
2. The method of drawing according to claim 1, further comprising:
when the first updating of the partial manuscript is finished by utilizing the second current output state information, displaying the updated partial manuscript so that the user modifies the updated partial manuscript;
if the modified updated partial manuscript is received, determining third current state information of the updated partial manuscript, and taking the third current state information as input of a trained preset network model to obtain fourth current output state information corresponding to the third current state information;
performing a second update on the updated partial manuscript after modification according to the fourth current output state information;
repeating the steps to automatically update until the user is satisfied;
and if the modified updated partial manuscript is not received, taking the second current output state information as the input of the trained preset network model, and continuing updating until the user is satisfied.
3. The drawing method according to claim 1, wherein said displaying the updated complete draft of the current drawing to confirm whether the user is satisfied with the complete draft comprises:
confirming whether the complete manuscript is a final manuscript;
if yes, optimizing the complete sketch and displaying the complete sketch, and acquiring satisfaction of the user;
otherwise, continuing to update the complete manuscript until the complete manuscript is determined to be the final manuscript.
4. A painting apparatus, the apparatus comprising:
the training module is used for training the preset network model by utilizing the preset drawing manuscript;
the first updating module is used for receiving partial manuscripts of the current picture drawn by the user and updating the partial manuscripts by utilizing the trained preset network model;
the confirmation module is used for displaying the updated complete manuscript so as to confirm whether the user is satisfied with the complete manuscript, wherein the updated complete manuscript comprises the updated partial manuscript;
the output module is used for outputting the complete manuscript after determining that the user is satisfied with the complete manuscript;
wherein, training module includes:
the first acquisition submodule is used for acquiring a preset number of preset drawings;
the second acquisition submodule is used for acquiring N images of each preset drawing at intervals of preset time or at preset steps in the process from the beginning of drawing to the end of drawing;
a first determining sub-module for determining first preset state information, second preset state information, and third preset state information of the N images from a first image to a last image;
the receiving sub-module is used for taking the first preset state information, the second preset state information and the third preset state information as input of the preset network model and receiving the first output state information, the second output state information and the third preset state information, wherein the first output state information, the second output state information and the third output state information are output by the preset network model;
the comparison sub-module is used for comparing the second preset state information with the first output state information to obtain first similarity, comparing the third preset state information with the second output state information to obtain second similarity, and repeating the steps to obtain N-1 similarities;
the training sub-module is used for training the preset network model by taking the N-1 similarities as a loss function;
wherein the first update module comprises:
the second determining submodule is used for determining first current state information of the partial drawing, wherein the first current state information is all information in a canvas area on a current drawing board;
a fourth obtaining sub-module, configured to input the first current state information into the trained preset network model to obtain second current output state information corresponding to the first current state information;
and the first updating sub-module is used for updating the partial manuscript for the first time according to the second current output state information.
5. The painting apparatus according to claim 4, wherein said apparatus further comprises:
the display module is used for displaying the updated partial manuscript so that the user modifies the updated partial manuscript when the first updating of the partial manuscript is finished by utilizing the second current output state information;
the determining module is used for determining third current state information of the updated partial manuscript if the updated partial manuscript is received, and taking the third current state information as input of a trained preset network model to obtain fourth current output state information corresponding to the third current state information;
the second updating module is used for updating the updated part of the manuscript after modification for the second time according to the fourth current output state information;
repeating the steps to automatically update until the user is satisfied;
and the third updating module is used for taking the second current output state information as the input of the trained preset network model and continuing updating until the user is satisfied if the modified updated partial draft is not received.
6. The painting apparatus according to claim 4, wherein the confirmation module comprises:
a confirmation sub-module for confirming whether the complete manuscript is a final manuscript;
the processing sub-module is used for carrying out optimization processing and displaying on the complete manuscript to obtain the satisfaction degree of the user if the complete manuscript is confirmed to be the final manuscript by the confirmation sub-module;
and the second updating sub-module is used for continuously updating the complete manuscript until the complete manuscript is determined to be the final manuscript if the confirmation sub-module confirms that the complete manuscript is not the final manuscript.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010147722.5A CN111311707B (en) | 2020-03-05 | 2020-03-05 | Painting method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010147722.5A CN111311707B (en) | 2020-03-05 | 2020-03-05 | Painting method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111311707A CN111311707A (en) | 2020-06-19 |
| CN111311707B true CN111311707B (en) | 2023-05-05 |
Family
ID=71162624
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010147722.5A Active CN111311707B (en) | 2020-03-05 | 2020-03-05 | Painting method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111311707B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108510456A (en) * | 2018-03-27 | 2018-09-07 | 华南理工大学 | The sketch of depth convolutional neural networks based on perception loss simplifies method |
| WO2019085655A1 (en) * | 2017-10-30 | 2019-05-09 | 上海寒武纪信息科技有限公司 | Information processing method and terminal device |
-
2020
- 2020-03-05 CN CN202010147722.5A patent/CN111311707B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019085655A1 (en) * | 2017-10-30 | 2019-05-09 | 上海寒武纪信息科技有限公司 | Information processing method and terminal device |
| CN108510456A (en) * | 2018-03-27 | 2018-09-07 | 华南理工大学 | The sketch of depth convolutional neural networks based on perception loss simplifies method |
Non-Patent Citations (1)
| Title |
|---|
| 图形图像软件对传统绘画画稿制作的介入;陆乐;;外语艺术教育研究(04);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111311707A (en) | 2020-06-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109190722B (en) | Font style migration transformation method based on Manchu character picture | |
| CN113240580A (en) | Lightweight image super-resolution reconstruction method based on multi-dimensional knowledge distillation | |
| CN109101545A (en) | Natural language processing method, apparatus, equipment and medium based on human-computer interaction | |
| CN106649746A (en) | Answer acquisition method and device | |
| CN108231066A (en) | Speech recognition system and method thereof and vocabulary establishing method | |
| CN114866807A (en) | Avatar video generation method and device, electronic equipment and readable storage medium | |
| CN112669215A (en) | Training text image generation model, text image generation method and device | |
| JP2019028094A (en) | Character generation device, program and character output device | |
| CN108921919A (en) | Animated show, production method and device | |
| CN111311707B (en) | Painting method and device | |
| CN111832530A (en) | Chemical formula identification system and method | |
| CN116909435A (en) | Data processing method and device, electronic equipment and storage medium | |
| CN107221019B (en) | Chart conversion method and device | |
| CN116433468A (en) | Data processing method and device for image generation | |
| CN111462263B (en) | Image generation method and device | |
| CN113450437A (en) | Ideological and political education propaganda system, data processing terminal and storage medium | |
| KR102584454B1 (en) | Mehtod of creating contents using big data machine learning and apparatus performing thereof | |
| US20240169701A1 (en) | Affordance-based reposing of an object in a scene | |
| CN116975344B (en) | Chinese character library generation method and device based on Stable diffration | |
| CN113420566B (en) | Method, device and medium for judging reasonability of long-term semantics | |
| US20240071132A1 (en) | Image annotation using prior model sourcing | |
| CN116052145A (en) | Correction method, correction device, storage medium and electronic equipment | |
| CN118212433A (en) | Method and device for processing national calligraphy teaching, electronic equipment and storage medium | |
| CN116842295A (en) | Front-end interface generation method, system, medium and terminal based on diffusion model | |
| CN117332769A (en) | Form generation method, system, electronic equipment and medium based on OCR technology |
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 |