CN109523460A - Moving method, moving apparatus and the computer readable storage medium of image style - Google Patents
Moving method, moving apparatus and the computer readable storage medium of image style Download PDFInfo
- Publication number
- CN109523460A CN109523460A CN201811269937.3A CN201811269937A CN109523460A CN 109523460 A CN109523460 A CN 109523460A CN 201811269937 A CN201811269937 A CN 201811269937A CN 109523460 A CN109523460 A CN 109523460A
- Authority
- CN
- China
- Prior art keywords
- image
- processing result
- content images
- style transfer
- target
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000003860 storage Methods 0.000 title claims abstract description 15
- 238000012545 processing Methods 0.000 claims abstract description 107
- 238000012546 transfer Methods 0.000 claims abstract description 75
- 230000011218 segmentation Effects 0.000 claims description 10
- 238000007796 conventional method Methods 0.000 abstract description 7
- 238000004891 communication Methods 0.000 description 9
- 238000010276 construction Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000013528 artificial neural network Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 4
- 238000013507 mapping Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 230000005236 sound signal Effects 0.000 description 4
- 230000015572 biosynthetic process Effects 0.000 description 3
- 238000003786 synthesis reaction Methods 0.000 description 3
- 230000000712 assembly Effects 0.000 description 2
- 238000000429 assembly Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000005012 migration Effects 0.000 description 2
- 238000013508 migration Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000002194 synthesizing effect Effects 0.000 description 2
- KLDZYURQCUYZBL-UHFFFAOYSA-N 2-[3-[(2-hydroxyphenyl)methylideneamino]propyliminomethyl]phenol Chemical compound OC1=CC=CC=C1C=NCCCN=CC1=CC=CC=C1O KLDZYURQCUYZBL-UHFFFAOYSA-N 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- 201000001098 delayed sleep phase syndrome Diseases 0.000 description 1
- 208000033921 delayed sleep phase type circadian rhythm sleep disease Diseases 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000010428 oil painting Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Classifications
-
- G06T3/04—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
Abstract
The application is moving method, moving apparatus and the computer storage medium about a kind of image style.The moving method includes: acquisition content images, wherein content images are to need to carry out the image of Local treatment in target area in Style Transfer;Target area is split from content images by target depth network, and Local treatment is carried out to target area, obtains the first processing result;Image of the content images after Style Transfer is obtained based on the first processing result.In the moving method, the purpose of Local treatment is individually carried out to target area by target depth network implementations, without complicating Style Transfer institute using network because of Local treatment, the more time-consuming technical problem of the used conventional method of image Style Transfer is alleviated.
Description
Technical field
The application belongs to technical field of image processing, especially image Style Transfer method and moving apparatus.
Background technique
Image stylization (image style transfer) is a kind of very interesting image processing method, in given one kind
Under the premise of image style, other images are transformed to the image of this style, image is such as made to realize oil painting stylizing, season
Stylization etc..Traditional learning method can be used in image stylization, and depth e-learning, depth network science also can be used
It practises and has more advantage in terms of accuracy.
Depth e-learning common at present includes carrying out deep learning by descriptive neural network and passing through production
Neural network carries out deep learning.Wherein, deep learning is carried out by descriptive neural network, that is, anti-since random noise
Realize that the stylization of content images, this kind of method minimize total losses to propagation iterative, so that the image after stylization simultaneously will
The content of content images is matched with the style of style image.Deep learning is carried out by production neural network, that is, is first instructed
Practice and save some common " style " parameters of user, then user's input needs to convert style picture into network, then specified net
The genre parameters that network calls finally export the output that picture is network the last layer, will not relate among network learning procedure
Excessive parameter adjustment and optimization, this method have very big promotion in calculating speed and calculating cost.
However, currently used above two depth online learning methods, if encountering figure during image stylization
As the case where partial analysis adjustment (for example, the construction window in the stylization of townscape night is lighted), then need sufficiently complex
Depth network be not suitable for the real-time place of video so that more time-consuming using the process of depth network stylization image
Reason.
Summary of the invention
To alleviate problem present in the relevant technologies, the application discloses a kind of image Style Transfer method and moving apparatus.
In a first aspect, the embodiment of the invention provides a kind of moving methods of image style, comprising:
Obtain content images, wherein the content images are to need to carry out at part in target area in Style Transfer
The image of reason;
The target area is split from the content images by target depth network, and to the target area
Domain carries out the Local treatment, obtains the first processing result;
Image of the content images after Style Transfer is obtained based on first processing result.
With reference to first aspect, the embodiment of the invention provides the first possible embodiments of first aspect, wherein
The target area is split from the content images by target depth network, and institute is carried out to the target area
When stating local directed complete set, Style Transfer processing is carried out to remaining area by target tool, obtains second processing result;
Image of the content images after Style Transfer is obtained based on first processing result, comprising: by described
First processing result and the second processing result synthesize image of the content images after Style Transfer, wherein
The remaining area is to remove remaining region behind the target area, the target tool in the content images
It is not the target depth network.
The possible embodiment of with reference to first aspect the first, the embodiment of the invention provides second of first aspect
Possible embodiment, wherein the content images are synthesized by first processing result and the second processing result and are passed through
Image after Style Transfer, comprising:
Adjustment is optimized to first processing result, obtains the adjustment result of first processing result;
The adjustment result and the second processing result are synthesized into image of the content images after Style Transfer.
With reference to first aspect, the embodiment of the invention provides the third possible embodiments of first aspect, wherein institute
Stating target depth network includes segmentation network.
Second aspect, the embodiment of the invention provides a kind of image Style Transfer devices, comprising:
Module is obtained, for obtaining content images, wherein the content images are to need in Style Transfer in target area
The image of domain progress Local treatment;
First processing module, for being partitioned into the target area from the content images by target depth network
Come, and the Local treatment is carried out to the target area, obtains the first processing result;
Determining module, for obtaining image of the content images after Style Transfer based on first processing result.
With reference to first aspect, the embodiment of the invention provides the first possible embodiments of first aspect, wherein moves
Moving device further include:
Second processing module, in the first processing module by target depth network by the target area from institute
When stating in content images and split, and carrying out the local directed complete set to the target area, by target tool to remaining area
Domain carries out Style Transfer processing, obtains second processing result;
The determining module is used for: synthesizing the content graph by first processing result and the second processing result
As the image after Style Transfer, wherein
The remaining area is to remove remaining region behind the target area, the target tool in the content images
It is not the target depth network.
The possible embodiment of with reference to first aspect the first, the embodiment of the invention provides second of first aspect
Possible embodiment, wherein the determining module is used for:
Adjustment is optimized to first processing result, obtains the adjustment result of first processing result;
The adjustment result and the second processing result are synthesized into image of the content images after Style Transfer.
With reference to first aspect, the embodiment of the invention provides the third possible embodiments of first aspect, wherein institute
Stating target depth network includes segmentation network.
The third aspect, the embodiment of the invention provides a kind of image Style Transfer devices, comprising:
Processor;
Memory for storage processor executable instruction;
Wherein, the processor is configured to executing moving method described in first aspect.
Fourth aspect, the embodiment of the invention provides a kind of computer readable storage medium, the computer-readable storage
Media storage has computer instruction, and the computer instruction, which is performed, realizes moving method described in first aspect.
The technical solution that embodiments herein provides can include the following benefits:
In the moving method, the mesh of Local treatment is individually carried out to target area by target depth network implementations
, without complicating Style Transfer using network because of Local treatment, alleviates image Style Transfer and use tradition
The more time-consuming technical problem of method.
Detailed description of the invention
Fig. 1 show a kind of moving method flow chart of image style of the offer of the embodiment of the present invention one;
Fig. 2 show a kind of moving method schematic diagram of image style of the offer of the embodiment of the present invention one;
Fig. 3 show a kind of moving method application example of image style of the offer of the embodiment of the present invention one;
Fig. 4 show a kind of moving apparatus structural block diagram of image style provided by Embodiment 2 of the present invention;
Fig. 5 show a kind of moving apparatus structural block diagram of image style of the offer of the embodiment of the present invention three;
Fig. 6 show the moving apparatus structural block diagram of another image style of the offer of the embodiment of the present invention three.
Specific embodiment
Example embodiments are described in detail here, and the example is illustrated in the accompanying drawings.Following description is related to
When attached drawing, unless otherwise indicated, the same numbers in different drawings indicate the same or similar elements.Following exemplary embodiment
Described in embodiment do not represent all embodiments consistent with the application.On the contrary, they be only with it is such as appended
The example of the consistent device and method of some aspects be described in detail in claims, the application.
For convenient for understanding the present embodiment, with reference to the accompanying drawings and examples, to a specific embodiment of the invention
It is described in further detail.
Embodiment one
The embodiment of the invention provides a kind of moving methods of image style, as shown in Figure 1, comprising:
Step S102 obtains content images, wherein content images are needed in Style Transfer in target area carry out office
The image of portion's processing;
Step S104 is split target area by target depth network from content images, and to target area
Local treatment is carried out, the first processing result is obtained;
Step S106 obtains image of the content images after Style Transfer based on the first processing result.
In the embodiment of the present invention, the mesh of Local treatment is individually carried out to target area by target depth network implementations
, without complicating Style Transfer using network because of Local treatment, alleviates image Style Transfer and use tradition
The more time-consuming technical problem of method.
In one optional embodiment of the embodiment of the present invention, in step S104, by target depth network by target area
Domain is split from content images, and to target area carry out local directed complete set when, by target tool to remaining area carry out
Style Transfer processing, obtains second processing result;
Step S106 obtains image of the content images after Style Transfer based on the first processing result, comprising: pass through first
The image of processing result and second processing result synthesis content images after Style Transfer, wherein
Remaining area is that remaining region behind target area is removed in content images, and target tool is not target depth net
Network.
Specifically, referring to Fig. 2, content images are loaded into respectively in target tool and target depth network;Then, target
Depth network is based on obtaining the first processing result after content images execute step S104, while target tool is based on content images pair
Remaining area carries out Style Transfer processing, obtains second processing result;Finally, passing through the first processing result and second processing result
Synthesize image of the content images after Style Transfer.
Townscape night stylization is as a kind of typical image Style Transfer process, it is usually required mainly for two processing,
In, one be full images color mapping, one is that construction window is lighted in image.Below with the stylization of townscape night
For, following image explanation is carried out to the embodiment of the present invention:
(1) content images are mostly color image, and color image generallys use RGB channel, are also possible to YUV or YCrCb etc.
Color space;
(2) target area is construction window, and Local treatment is to light to construction window, and the first processing result is to light
Construction window;
(3) remaining area is that remaining region after construction window is removed in content images, and Style Transfer processing is will be remaining
Area image content carries out color mapping, and second processing result is that the picture material at night is in remaining area;
(4) target tool is depth network or traditional color mapping (color mapping) algorithm, it is emphasized that
It is that target tool is not target depth network, that is, when target tool is depth network, depth network used in target tool
It is not the same depth network with depth network used in step S104.
It is application example of the moving method of image style in the stylization of townscape night shown in Fig. 3, wherein figure
A is content images, and figure B1 is second processing as a result, figure B2 is the first processing result, and figure C is content images after Style Transfer
Image.
In the embodiment of the present invention, target tool is not target depth network, that is, the first processing result and second processing result
Obtain process be two parallel operational process of tool, more time saving, the transition process of image style is quicker, further delays
The more time-consuming technical problem of the used conventional method of image Style Transfer is solved.
In another optional embodiment of the embodiment of the present invention, synthesized by the first processing result and second processing result
Image of the content images after Style Transfer, comprising:
Adjustment is optimized to the first processing result, obtains the adjustment result of the first processing result;
The image of result and second processing result synthesis content images after Style Transfer will be adjusted.
It should be noted that adjustment is optimized to the first processing result, in order to which content images are after Style Transfer
Image is more life-like and is adjusted to the first processing result.For the stylization of townscape night, i.e., to the first processing result
Brightness is adjusted, changes color, change detection ratio etc., so that image of the content images after Style Transfer is more cut
Nearly night shows image.
In the embodiment of the present invention, by optimizing adjustment to the first processing result, realize to content image local area
The controllable purpose in domain, it is more controllable to advantageously allow image of the content images after Style Transfer, so that content images are through style
Image after migration reaches expected.
In another optional embodiment of the embodiment of the present invention, target depth network includes segmentation network.
It should be noted that segmentation network is the depth network with picture segmentation function.Segmentation picture may be implemented
Under the premise of function, the depth network the simple more advantageously reduces the used time of Style Transfer method.
Specifically, after segmentation network is added in content images, the target for dividing the network operation is before being divided into content images
Scape and background.In the stylization of townscape night, prospect is construction window, and background is to remain after content images remove construction window
Remaining scenery.
In the embodiment of the present invention, by simply dividing network, so that it may obtain target area, operational process is simply fast
Victory is conducive to alleviate the more time-consuming technical problem of the used conventional method of image Style Transfer.
Embodiment two
The embodiment of the present invention provides a kind of image Style Transfer device, as shown in Figure 4, comprising:
Module 100 is obtained, for obtaining content images, wherein content images are to need in Style Transfer in target area
The image of domain progress Local treatment;
First processing module 200, for target area to be split from content images by target depth network, and
Local treatment is carried out to target area, obtains the first processing result;
Determining module 300, for obtaining image of the content images after Style Transfer based on the first processing result.
Image Style Transfer device provided in an embodiment of the present invention, by target depth network implementations individually to target area
Domain carries out the purpose of Local treatment, without complicating Style Transfer institute using network because of Local treatment, alleviates image
The more time-consuming technical problem of the used conventional method of Style Transfer.
In one optional embodiment of the embodiment of the present invention, moving apparatus further include:
Second processing module, for first processing module by target depth network by target area from content images
When splitting, and carrying out local directed complete set to target area, Style Transfer processing is carried out to remaining area by target tool, is obtained
To second processing result;
Determining module is used for: synthesizing content images after Style Transfer by the first processing result and second processing result
Image, wherein
Remaining area is that remaining region behind target area is removed in content images, and target tool is not target depth net
Network.
In another optional embodiment of the embodiment of the present invention, determining module is used for:
Adjustment is optimized to the first processing result, obtains the adjustment result of the first processing result;
The image of result and second processing result synthesis content images after Style Transfer will be adjusted.
In another optional embodiment of the embodiment of the present invention, target depth network includes segmentation network.
Embodiment three
The embodiment of the invention provides a kind of image Style Transfer devices, comprising:
Processor;
Memory for storage processor executable instruction;
Wherein, processor is configured as executing the moving method of embodiment one.
In the embodiment of the present invention, processor is configured as executing the moving method of embodiment one, that is, content images are obtained,
Wherein, content images are to need to carry out the image of Local treatment in target area in Style Transfer;Pass through target depth network
Target area is split from content images, and Local treatment is carried out to target area, obtains the first processing result;It is based on
First processing result obtains image of the content images after Style Transfer.It is real by target depth network in the moving method
The purpose that Local treatment is individually carried out to target area is showed, without answering Style Transfer institute using network because of Local treatment
Hydridization alleviates the more time-consuming technical problem of the used conventional method of image Style Transfer.
Fig. 5 show a kind of structural block diagram of moving apparatus 500.Referring to Fig. 5, moving apparatus 500 may include with next
A or multiple components: processing component 502, memory 504, electric power assembly 505, multimedia component 508, audio component 510 are defeated
Enter/export (I/O) interface 512, sensor module 514 and communication component 515.
The integrated operation of the usually control moving apparatus 500 of processing component 502, is such as grasped with display, data communication, and record
Make associated operation.Processing component 502 may include one or more processors 520 to execute instruction, above-mentioned to complete
The all or part of the steps of method.In addition, processing component 502 may include one or more modules, it is convenient for 502 He of processing component
Interaction between other assemblies.For example, processing component 502 may include multi-media module, to facilitate multimedia component 508 and place
Manage the interaction between component 502.
Memory 504 is configured as storing various types of data to support the operation in moving apparatus 500.These data
Example include any application or method for operating on device 500 instruction, contact data, telephone directory number
According to, message, picture, video etc..Memory 504 can by any kind of volatibility or non-volatile memory device or they
Combination realize, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM) is erasable
Programmable read only memory (EPROM), programmable read only memory (PROM), read-only memory (ROM), magnetic memory, quick flashing
Memory, disk or CD.
Power supply module 505 provides electric power for the various assemblies of device 500.Power supply module 505 may include power management system
System, one or more power supplys and other with for moving apparatus 500 generate, manage, and distribute the associated component of electric power.
Multimedia component 508 includes the screen of one output interface of offer between device 500 and user.In some realities
It applies in example, screen may include liquid crystal display (LCD) and touch panel (TP).If screen includes touch panel, screen can
To be implemented as touch screen, to receive input signal from the user.Touch panel include one or more touch sensors with
Sense the gesture on touch, slide, and touch panel.Touch sensor can not only sense the boundary of a touch or slide action, and
And also detect duration and pressure relevant to touch or slide.In some embodiments, multimedia component 508 includes
One front camera and/or rear camera.It is such as in a shooting mode or a video mode, preceding when equipment 500 is in operation mode
It sets camera and/or rear camera can receive external multi-medium data.Each front camera and rear camera can
Be a fixed optical lens system or have focusing and optical zoom capabilities.
Audio component 510 is configured as output and/or input audio signal.For example, audio component 510 includes a Mike
Wind (MIC), when moving apparatus 500 is in operation mode, when such as call mode, recording mode, and voice recognition mode, microphone
It is configured as receiving external audio signal.The received audio signal can be further stored in memory 504 or via logical
Believe that component 515 is sent.In some embodiments, audio component 510 further includes a loudspeaker, is used for output audio signal.
I/O interface 512 provides interface between processing component 502 and peripheral interface module, and above-mentioned peripheral interface module can
To be keyboard, click wheel, button etc..These buttons may include, but are not limited to: home button, volume button, start button and lock
Determine button.
Sensor module 514 includes one or more sensors, and the state for providing various aspects for device 500 is commented
Estimate.For example, sensor module 514 can detecte the state that opens/closes of equipment 500, the relative positioning of component, such as component
For the display and keypad of moving apparatus 500, sensor module 514 can also detect moving apparatus 500 or moving apparatus 500
The position change of one component, the existence or non-existence that user contacts with moving apparatus 500,500 orientation of moving apparatus or acceleration/
The temperature change slowed down with moving apparatus 500.Sensor module 514 may include proximity sensor, be configured to do not having
It is detected the presence of nearby objects when any physical contact.Sensor module 514 can also include optical sensor, such as CMOS or
Ccd image sensor, for being used in imaging applications.In some embodiments, which can also include adding
Velocity sensor, gyro sensor, Magnetic Sensor, pressure sensor or temperature sensor.
Communication component 515 is configured to facilitate the communication of wired or wireless way between moving apparatus 500 and other equipment.
Moving apparatus 500 can access the wireless network based on communication standard, such as WiFi, carrier network (such as 2G, 3G, 4G or 5G),
Or their combination.In one exemplary embodiment, communication component 515 receives via broadcast channel and comes from external broadcasting management
The broadcast singal or broadcast related information of system.In one exemplary embodiment, communication component 515 further includes near-field communication
(NFC) module, to promote short range communication.For example, radio frequency identification (RFID) technology, Infrared Data Association can be based in NFC module
(IrDA) technology, ultra wide band (UWB) technology, bluetooth (BT) technology and other technologies are realized.
In the exemplary embodiment, moving apparatus 500 can be by one or more application specific integrated circuit (ASIC), number
Word signal processor (DSP), digital signal processing appts (DSPD), programmable logic device (PLD), field programmable gate array
(FPGA), controller, microcontroller, microprocessor or other electronic components are realized, for executing the above method.
Fig. 6 show the structural block diagram of another moving apparatus 600.For example, moving apparatus 600 may be provided as a clothes
Business device.Referring to Fig. 6, it further comprises one or more processors, and by depositing that moving apparatus 600, which includes processing component 622,
Memory resource representated by reservoir 632, can be by the instruction of the execution of processing component 622, such as application program for storing.It deposits
The application program stored in reservoir 632 may include it is one or more each correspond to one group of instruction module.This
Outside, processing component 622 is configured as executing instruction, to execute above- mentioned information list display method.
Moving apparatus 600 can also include that a power supply module 625 is configured as executing the power supply pipe of moving apparatus 600
Reason, a wired or wireless network interface 650 are configured as moving apparatus 600 being connected to network and an input and output (I/
O) interface 658.Moving apparatus 600 can be operated based on the operating system for being stored in memory 632, such as Windows
ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM or similar.
Example IV
The embodiment of the invention provides a kind of computer readable storage medium, computer-readable recording medium storage has calculating
Machine instruction, computer instruction are performed the moving method for realizing embodiment one.
Specifically, computer readable storage medium, the memory 504 for example including instruction, above-metioned instruction can be filled by migration
Set 500 processor 520 execute to complete the above method.For example, non-transitorycomputer readable storage medium can be ROM,
Random access memory (RAM), CD-ROM, tape, floppy disk and optical data storage devices etc..
In the embodiment of the present invention, computer instruction is performed the moving method for realizing embodiment one, that is, obtains content graph
Picture, wherein content images are to need to carry out the image of Local treatment in target area in Style Transfer;Pass through target depth net
Network splits target area from content images, and carries out Local treatment to target area, obtains the first processing result;Base
Image of the content images after Style Transfer is obtained in the first processing result.In the moving method, pass through target depth network
Realize the purpose that Local treatment is individually carried out to target area, without because Local treatment and by the used network of Style Transfer
It complicates, alleviates the more time-consuming technical problem of the used conventional method of image Style Transfer.
Those skilled in the art after considering the specification and implementing the invention disclosed here, will readily occur to its of the application
Its embodiment.This application is intended to cover any variations, uses, or adaptations of the application, these modifications, purposes or
Person's adaptive change follows the general principle of the application and including the undocumented common knowledge in the art of the application
Or conventional techniques.The description and examples are only to be considered as illustrative, and the true scope and spirit of the application are by following
Claim is pointed out.
It should be understood that the application is not limited to the precise structure that has been described above and shown in the drawings, and
And various modifications and changes may be made without departing from the scope thereof.Scope of the present application is only limited by the accompanying claims.
Claims (10)
1. a kind of moving method of image style characterized by comprising
Obtain content images, wherein the content images are to need to carry out Local treatment in target area in Style Transfer
Image;
The target area is split from the content images by target depth network, and to the target area into
The row Local treatment, obtains the first processing result;
Image of the content images after Style Transfer is obtained based on first processing result.
2. moving method according to claim 1, which is characterized in that
The target area is being split from the content images by target depth network, and to the target area
When carrying out the local directed complete set, Style Transfer processing is carried out to remaining area by target tool, obtains second processing result;
Image of the content images after Style Transfer is obtained based on first processing result, comprising: passes through described first
Processing result and the second processing result synthesize image of the content images after Style Transfer, wherein
The remaining area is to remove remaining region behind the target area in the content images, and the target tool is not
The target depth network.
3. moving method according to claim 2, which is characterized in that at first processing result and described second
Reason result synthesizes image of the content images after Style Transfer, comprising:
Adjustment is optimized to first processing result, obtains the adjustment result of first processing result;
The adjustment result and the second processing result are synthesized into image of the content images after Style Transfer.
4. moving method according to claim 1, which is characterized in that the target depth network includes segmentation network.
5. a kind of image Style Transfer device characterized by comprising
Obtain module, for obtaining content images, wherein the content images be needed in Style Transfer target area into
The image of row Local treatment;
First processing module, for the target area to be split from the content images by target depth network,
And the Local treatment is carried out to the target area, obtain the first processing result;
Determining module, for obtaining image of the content images after Style Transfer based on first processing result.
6. moving apparatus according to claim 5, which is characterized in that further include:
Second processing module, in the first processing module by target depth network by the target area from described interior
Hold image in split, and to the target area carry out the local directed complete set when, by target tool to remaining area into
The processing of row Style Transfer, obtains second processing result;
The determining module is used for: being synthesized the content images by first processing result and the second processing result and is passed through
Image after Style Transfer, wherein
The remaining area is to remove remaining region behind the target area in the content images, and the target tool is not
The target depth network.
7. moving apparatus according to claim 6, which is characterized in that the determining module is used for:
Adjustment is optimized to first processing result, obtains the adjustment result of first processing result;
The adjustment result and the second processing result are synthesized into image of the content images after Style Transfer.
8. moving apparatus according to claim 5, which is characterized in that the target depth network includes segmentation network.
9. a kind of image Style Transfer device characterized by comprising
Processor;
Memory for storage processor executable instruction;
Wherein, the processor is configured to executing moving method described in the claims 1-4 any one.
10. a kind of computer readable storage medium, which is characterized in that the computer-readable recording medium storage has computer to refer to
It enables, the computer instruction is performed realization such as the described in any item moving methods of Claims 1-4.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811269937.3A CN109523460A (en) | 2018-10-29 | 2018-10-29 | Moving method, moving apparatus and the computer readable storage medium of image style |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811269937.3A CN109523460A (en) | 2018-10-29 | 2018-10-29 | Moving method, moving apparatus and the computer readable storage medium of image style |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109523460A true CN109523460A (en) | 2019-03-26 |
Family
ID=65773086
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811269937.3A Pending CN109523460A (en) | 2018-10-29 | 2018-10-29 | Moving method, moving apparatus and the computer readable storage medium of image style |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109523460A (en) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110264413A (en) * | 2019-05-17 | 2019-09-20 | 北京达佳互联信息技术有限公司 | A kind of image processing method, device, electronic equipment and storage medium |
CN110310222A (en) * | 2019-06-20 | 2019-10-08 | 北京奇艺世纪科技有限公司 | A kind of image Style Transfer method, apparatus, electronic equipment and storage medium |
CN110689478A (en) * | 2019-09-25 | 2020-01-14 | 北京字节跳动网络技术有限公司 | Image stylization processing method and device, electronic equipment and readable medium |
CN111223039A (en) * | 2020-01-08 | 2020-06-02 | 广东博智林机器人有限公司 | Image style conversion method and device, electronic equipment and storage medium |
CN111415299A (en) * | 2020-03-26 | 2020-07-14 | 浙江科技学院 | High-resolution image style migration method |
CN111429342A (en) * | 2020-03-31 | 2020-07-17 | 河南理工大学 | Photo style migration method based on style corpus constraint |
CN111950608A (en) * | 2020-06-12 | 2020-11-17 | 中国科学院大学 | Domain self-adaptive object detection method based on contrast loss |
CN111986076A (en) * | 2020-08-21 | 2020-11-24 | 深圳市慧鲤科技有限公司 | Image processing method and device, interactive display device and electronic equipment |
CN113160039A (en) * | 2021-04-28 | 2021-07-23 | 北京达佳互联信息技术有限公司 | Image style migration method and device, electronic equipment and storage medium |
CN113160033A (en) * | 2020-12-28 | 2021-07-23 | 武汉纺织大学 | Garment style migration system and method |
CN113469876A (en) * | 2021-07-28 | 2021-10-01 | 北京达佳互联信息技术有限公司 | Image style migration model training method, image processing method, device and equipment |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102737369A (en) * | 2011-03-31 | 2012-10-17 | 卡西欧计算机株式会社 | Image processing apparatus, image processing method, and storage medium |
CN106778928A (en) * | 2016-12-21 | 2017-05-31 | 广州华多网络科技有限公司 | Image processing method and device |
CN108205813A (en) * | 2016-12-16 | 2018-06-26 | 微软技术许可有限责任公司 | Image stylization based on learning network |
CN108550107A (en) * | 2018-04-27 | 2018-09-18 | Oppo广东移动通信有限公司 | A kind of image processing method, picture processing unit and mobile terminal |
-
2018
- 2018-10-29 CN CN201811269937.3A patent/CN109523460A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102737369A (en) * | 2011-03-31 | 2012-10-17 | 卡西欧计算机株式会社 | Image processing apparatus, image processing method, and storage medium |
CN108205813A (en) * | 2016-12-16 | 2018-06-26 | 微软技术许可有限责任公司 | Image stylization based on learning network |
CN106778928A (en) * | 2016-12-21 | 2017-05-31 | 广州华多网络科技有限公司 | Image processing method and device |
CN108550107A (en) * | 2018-04-27 | 2018-09-18 | Oppo广东移动通信有限公司 | A kind of image processing method, picture processing unit and mobile terminal |
Cited By (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11328399B2 (en) | 2019-05-17 | 2022-05-10 | Beijing Daja Internet Information Technology Co., Ltd. | Method and apparatus, and storage medium for processing style image |
CN110264413A (en) * | 2019-05-17 | 2019-09-20 | 北京达佳互联信息技术有限公司 | A kind of image processing method, device, electronic equipment and storage medium |
CN110310222A (en) * | 2019-06-20 | 2019-10-08 | 北京奇艺世纪科技有限公司 | A kind of image Style Transfer method, apparatus, electronic equipment and storage medium |
CN110689478A (en) * | 2019-09-25 | 2020-01-14 | 北京字节跳动网络技术有限公司 | Image stylization processing method and device, electronic equipment and readable medium |
CN110689478B (en) * | 2019-09-25 | 2023-12-01 | 北京字节跳动网络技术有限公司 | Image stylization processing method and device, electronic equipment and readable medium |
CN111223039A (en) * | 2020-01-08 | 2020-06-02 | 广东博智林机器人有限公司 | Image style conversion method and device, electronic equipment and storage medium |
CN111415299A (en) * | 2020-03-26 | 2020-07-14 | 浙江科技学院 | High-resolution image style migration method |
CN111415299B (en) * | 2020-03-26 | 2023-03-28 | 浙江科技学院 | High-resolution image style migration method |
CN111429342A (en) * | 2020-03-31 | 2020-07-17 | 河南理工大学 | Photo style migration method based on style corpus constraint |
CN111429342B (en) * | 2020-03-31 | 2024-01-05 | 河南理工大学 | Photo style migration method based on style corpus constraint |
CN111950608B (en) * | 2020-06-12 | 2021-05-04 | 中国科学院大学 | Domain self-adaptive object detection method based on contrast loss |
CN111950608A (en) * | 2020-06-12 | 2020-11-17 | 中国科学院大学 | Domain self-adaptive object detection method based on contrast loss |
CN111986076A (en) * | 2020-08-21 | 2020-11-24 | 深圳市慧鲤科技有限公司 | Image processing method and device, interactive display device and electronic equipment |
CN113160033A (en) * | 2020-12-28 | 2021-07-23 | 武汉纺织大学 | Garment style migration system and method |
CN113160039A (en) * | 2021-04-28 | 2021-07-23 | 北京达佳互联信息技术有限公司 | Image style migration method and device, electronic equipment and storage medium |
CN113160039B (en) * | 2021-04-28 | 2024-03-26 | 北京达佳互联信息技术有限公司 | Image style migration method and device, electronic equipment and storage medium |
CN113469876A (en) * | 2021-07-28 | 2021-10-01 | 北京达佳互联信息技术有限公司 | Image style migration model training method, image processing method, device and equipment |
CN113469876B (en) * | 2021-07-28 | 2024-01-09 | 北京达佳互联信息技术有限公司 | Image style migration model training method, image processing method, device and equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109523460A (en) | Moving method, moving apparatus and the computer readable storage medium of image style | |
CN105100388B (en) | Mobile terminal and its control method | |
CN104902189A (en) | Picture processing method and picture processing device | |
CN110288084A (en) | Super-network training method and device | |
CN108986199A (en) | Dummy model processing method, device, electronic equipment and storage medium | |
CN108762839A (en) | The interface display method and device of application program | |
CN107368810A (en) | Method for detecting human face and device | |
CN108037863A (en) | A kind of method and apparatus for showing image | |
CN103207728A (en) | Method Of Providing Augmented Reality And Terminal Supporting The Same | |
CN105959546A (en) | Panorama shooting device and method | |
CN106603917A (en) | Shooting device and method | |
WO2022198934A1 (en) | Method and apparatus for generating video synchronized to beat of music | |
CN104463092B (en) | Mobile terminal and its control method | |
CN104133956A (en) | Method and device for processing pictures | |
WO2020103353A1 (en) | Multi-beam selection method and device | |
KR102545408B1 (en) | Method and apparatus for detecting occluded image and medium | |
KR20160127606A (en) | Mobile terminal and the control method thereof | |
CN107341777A (en) | image processing method and device | |
CN109257649B (en) | Multimedia file generation method and terminal equipment | |
CN107395982A (en) | Photographic method and device | |
CN107291772A (en) | One kind search access method, device and electronic equipment | |
CN108965770A (en) | Image processing template generation method, device, storage medium and mobile terminal | |
CN108769769A (en) | Playback method, device and the computer readable storage medium of video | |
CN107566750A (en) | Control method, device and the storage medium of flash lamp | |
CN105426904B (en) | Photo processing method, device and equipment |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190326 |