CN115546338A - Image coloring method based on Transformer and generation countermeasure network - Google Patents
Image coloring method based on Transformer and generation countermeasure network Download PDFInfo
- Publication number
- CN115546338A CN115546338A CN202211247125.5A CN202211247125A CN115546338A CN 115546338 A CN115546338 A CN 115546338A CN 202211247125 A CN202211247125 A CN 202211247125A CN 115546338 A CN115546338 A CN 115546338A
- Authority
- CN
- China
- Prior art keywords
- color image
- image
- network
- vector
- transformer
- 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
- 238000004040 coloring Methods 0.000 title claims abstract description 45
- 238000000034 method Methods 0.000 title claims abstract description 26
- 230000008569 process Effects 0.000 claims abstract description 7
- 230000006870 function Effects 0.000 claims description 23
- 238000010606 normalization Methods 0.000 claims description 19
- 230000007246 mechanism Effects 0.000 claims description 13
- 238000004422 calculation algorithm Methods 0.000 claims description 9
- 238000009877 rendering Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 8
- 230000000694 effects Effects 0.000 claims description 7
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 claims description 3
- 230000009471 action Effects 0.000 claims description 3
- 238000013527 convolutional neural network Methods 0.000 abstract description 6
- 238000012549 training Methods 0.000 abstract description 3
- 230000000007 visual effect Effects 0.000 abstract 1
- 238000013461 design Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 5
- 230000004913 activation Effects 0.000 description 3
- 238000003909 pattern recognition Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/001—Texturing; Colouring; Generation of texture or colour
-
- 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
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T15/00—3D [Three Dimensional] image rendering
- G06T15/005—General purpose rendering architectures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/80—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level
- G06V10/806—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Multimedia (AREA)
- Medical Informatics (AREA)
- Databases & Information Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Computer Graphics (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method for coloring images based on a Transformer and a generation countermeasure network, which solves the problem of coloring images by using the generation countermeasure network GAN and the Transformer instead of simply using a convolutional neural network CNN. The local enhanced forward propagation network and the hopping connection ensure that shallow features can be efficiently transmitted and utilized in the network so that the transform-GAN can efficiently capture the correlation between global and local information. An optimal training process is also explored through data enhancement and objective function selection, and a color image generator and a discriminator are formed to enable the Transformer-GAN to perform well in the aspect of image colorization. The best visual effect is achieved.
Description
Technical Field
The invention belongs to the technical field of image processing, and relates to an image coloring method based on a Transformer and a generation countermeasure network.
Background
In the image rendering task, our goal is to generate a color image from an input grayscale image. By category, from earlier conventional non-hopping connection algorithms based on CNN structures, networks of user-specified image colors (which require the user to enter color values in a particular layer) have emerged in the future. And end-to-end feed-forward architecture using animated image colorization that generates a countermeasure network (GAN), in addition to infrared colorization, radar image colorization, etc. for specific areas, and multi-modal coloring models (text-based coloring networks) that emerge later. Diversified colorization networks compensate for the lack of diversity by generating different color images, network architectures that incorporate multi-path networks that learn different characteristics in different network paths or levels, and users give reference images as input samples for the colored network. All the above models have in common that they are networks based on convolutional neural networks CNN, however unlike previous work i constructed image colorization networks using transformers and generation of confrontation networks GAN, which, to our knowledge, was the first study of image colorization using transformers as the main network.
Disclosure of Invention
The invention aims to provide an image coloring method based on a Transformer and a generation countermeasure network, which solves the problems of poor coloring effect and poor coloring diversity of the conventional image coloring network.
The technical scheme adopted by the invention is that the image coloring method based on the Transformer and the generation countermeasure network is implemented according to the following steps:
step 1, constructing an image coloring model based on a generation countermeasure network, wherein the image coloring model comprises a color image generator and a discriminator; the color image generator is used for generating a color image, and the discriminator is used for judging whether the input image is a real color image or a false color image;
step 2, inputting the gray image into a color image generator of the image coloring model to generate a pseudo color image;
and 3, respectively updating parameters of the discriminator and the color image generator:
step 3.1: firstly, fixing parameters of a color image generator, inputting the false color image and the real color image corresponding to the gray image into an identifier in sequence, then calculating the loss between the real color image corresponding to the gray image and a label value of 1 according to a loss function, calculating the loss between the false color image generated by the gray image and a label value of 0 according to the loss function, and finally updating the parameters of the identifier by using a back propagation algorithm; wherein a label value of 1 represents a real image and a label value of 0 represents a generated pseudo color image;
step 3.2: and fixing parameters of the discriminator, calculating the loss between the generated pseudo color image and the label value of 1 according to a loss function, and finally updating the parameters of the color image generator by utilizing a back propagation algorithm.
Step 3.3: continuously circulating the process of updating parameters of the discriminator and the color image generator in the steps 3.1 and 3.2 until the loss value is converged, and generating a false color image with good effect by the color image generator, namely obtaining an optimized image coloring model;
and 4, directly coloring the gray image by using the optimized image coloring model.
The present invention is also characterized in that,
in step 1, the color image generator includes a plurality of MWin-transformer modules, where the MWin-transformer modules are used to extract and reconstruct features of an image, and output a 3-channel effective color image:
the Mwin-transformer module consists of three core parts: a window-based multi-head self-attention mechanism, a layer normalization operation LN, and a local enhanced forward propagation network LeFF.
The flow of the color image generator to generate a pseudo color image is as follows:
X′＝Embedded Tokens(X in )
X″＝W-MSA(LN(X))+X′
X out ＝LeFF(LN(X″))+X″
wherein, X in Representing an input as a gray image or a pseudo-color image;
embedding Tokens represents dividing X in Converting into a vector;
x' represents X in Inputting the vector obtained by Embedding Tokens and outputting the vector;
then inputting a result LN (X ') obtained by carrying out layer normalization on the vector X' into a multi-head self-attention mechanism W-MSA based on a window to obtain a vector with extracted characteristic information, and adding the vector with X 'to obtain a vector X' with more collected characteristic information; x 'represents the input of X' into the output of the window-based multi-headed autofocusing mechanism and layer normalization operation;
continuously carrying out layer normalization on the vector X ', inputting the normalized LN (X ') into a local enhanced forward propagation network to obtain a vector with more extracted local characteristic information, and adding the vector with X ' to obtain a vector X with more gathered local characteristic information out ，X out Represents the input of X "into the locally enhanced forward propagation network LeFF and the resulting output of the layer normalization operation.
The layer normalization LN operation is to solve the problem of internal covariate offset, and the calculation process of the layer normalization operation is as follows:
wherein, the action object of LN layer is
X represents a vector, μ and δ represent the mean and variance of each sample,
and
as affine learning parameters, d k Is the dimension of the concealment that is in place,
indicating that the number is a k-dimensional vector.
The window-based multi-head self-attention mechanism is as follows:
the false color image is divided into a plurality of windows, then self-attention calculation is carried out in the different windows, and as the patch number in one window is far less than all small blocks in one image and the number of the windows is kept unchanged, the calculation complexity of the multi-head self-attention mechanism based on the windows and the image size are changed into a linear relation from a square relation, and the calculation complexity of the model is greatly reduced.
And adding convolution to a forward propagation network in the Mwin-transformer module so as to form a local enhanced forward propagation network LeFF.
The loss function is:
wherein,
wherein G is * The sum of the loss functions is represented,
the presentation condition is generated against the network loss,
denotes the Charbonier loss, λ denotes the weighting coefficient of the Charbonier loss;
x represents an input gray image;
y represents a real color image corresponding to the input gray image;
log represents a base 2 logarithmic function;
ε represents a value of 10 -3 Constant coefficient of (d);
and | | represents solving an absolute value.
The invention has the beneficial effects that: the invention relates to a method for image coloring based on a Transformer and a generation countermeasure network, which can be used for coloring gray images. The image coloring method has good coloring effect on the gray image in detail or the whole, is suitable for the gray image with any size, and has high universality.
Drawings
FIG. 1 is a block diagram of an image rendering model of the present invention;
fig. 2 is a structural diagram of the color image generator G;
FIG. 3 (a) is a block diagram of an discriminator D according to the invention;
FIG. 3 (b) is a structural diagram of the MWin-transporter of the present invention;
fig. 3 (c) is a structural diagram of the local enhanced forward propagation network LeFF of the present invention.
Detailed Description
The invention discloses an image coloring method based on a Transformer and a generation countermeasure network, which is implemented according to the following steps:
step 1, constructing an image coloring model based on a generation countermeasure network, wherein the image coloring model comprises a color image generator and a discriminator; the color image generator is used for generating a color image, and the discriminator is used for judging whether the input image is a real color image or a false color image;
step 2, inputting the gray image into a color image generator of the image coloring model to generate a pseudo color image;
and 3, respectively updating parameters of the discriminator and the color image generator:
step 3.1: firstly, fixing parameters of a color image generator, sequentially and alternately inputting the pseudo color image and a real color image corresponding to the gray image into a discriminator, then calculating the loss between the real color image corresponding to the gray image and a label value of 1 according to a loss function, calculating the loss between the pseudo color image generated by the gray image and a label value of 0 according to the loss function, and finally updating the parameters of the discriminator by utilizing a back propagation algorithm; wherein a label value of 1 represents a real image and a label value of 0 represents a generated pseudo color image;
step 3.2: and fixing parameters of the discriminator, calculating the loss between the generated pseudo color image and the label value of 1 according to a loss function, and finally updating the parameters of the color image generator by using a back propagation algorithm.
Step 3.3: continuously circulating the process of updating parameters of the discriminator and the color image generator in the steps 3.1 and 3.2 until the loss value is converged, and generating a false color image with good effect by the color image generator, namely obtaining an optimized image coloring model;
and 4, directly coloring the gray image by using the optimized image coloring model.
As shown in FIG. 1, an image coloring model based on Transformer and generation countermeasure network GAN is constructed, G and D represent a color image generator and a discriminator respectively, and specifically, a gray image x ∈ R 3×H×W As input to the color image generator G, a pseudo-color image G (x) is generated, and then the pseudo-color image G (x) and the true-color image y are alternately input to the discriminator.
Firstly, fixing parameters of a color image generator, sequentially and alternately inputting the pseudo color image and a real color image corresponding to the gray image into a discriminator, then calculating the loss between the real color image corresponding to the gray image and a label value of 1 according to a loss function, calculating the loss between the pseudo color image generated by the gray image and a label value of 0 according to the loss function, and finally updating the parameters of the discriminator by utilizing a back propagation algorithm; where a label value of 1 represents a real image and a label value of 0 represents a generated pseudo color image.
And fixing parameters of the discriminator, calculating the loss between the generated pseudo color image and the label value of 1 according to a loss function, and finally updating the parameters of the color image generator by utilizing a back propagation algorithm. And (3) continuously circulating the process of updating the parameters of the discriminator and the color image generator in the steps 3.1 and 3.2 until the loss value is converged, and generating a false color image with good effect by the color image generator, namely obtaining an optimized image coloring model. And directly coloring the gray image by using the optimized image coloring model.
The design of the method mainly comprises two main points, namely the design of the color image generator and the discriminator and the design of the components thereof, and the detailed construction of the color image generator and the discriminator is described one by one below.
(1) Design of color image generator G:
the input and output of image colorization is a mapping relationship, and the depth conversion process of these components should be a symmetric relationship, based on which, as shown in fig. 2, the whole color image generator is designed into a U shape; in general, at the encoder stage, the gray image x is first subjected to input dimension adjustment, which is a convolution layer of convolution kernel 3 × 3, and the activation function LeakyReLU is used to adjust the input dimensions and extract low-level features. Then, after the designed window-based transformer module MWin-transformer, a downsampling layer consisting of convolution layers with a convolution kernel of step 2 by 4 is reached, and this step is repeated 2 times. The image is then passed through the MWin-transformer module as the bottleneck stage. The decoder and the encoder correspond to each other, and the design of the decoder and the encoder is completely symmetrical and consistent: first, it passes through the window MWin-transformer module and performs upsampling, where the upsampling operation is a transposed convolution with 2 x 2 of the convolution kernel with step size of 2. This step is repeated 2 times as the encoder in order to maintain the symmetry of the network. Finally, the output dimensions are adjusted using an output dimension adjustment consisting of a 3 x 3 convolution to ensure that the output is a 3-channel valid color image.
(2) Design of MWin-transform module
In FIG. 3 (b), the MWin-transformer module was constructed, consisting of three core parts: a W-MHSA mechanism, a layer normalization operation LN and a local enhanced forward propagation network LeFF network;
wherein the layer normalization operation LN is as follows:
the LN layer is an important guarantee for fast training and stable convergence of the image coloring model, and the LN layer is used as an action object
And
the mean and variance of each sample are represented separately,
and
as affine learning parameters, d k Is the hidden dimension. The calculation flow of the MWin-transformer module is as follows:
X′＝Embedded Tokens(X in )
X″＝W-MSA(LN(X))+X′
X out ＝LeFF(LN(X″))+X ″
wherein, X in Indicating inputIn, it is gray image or false color image;
embedding token Embedding Tokens denotes to assign X to in Converting into a vector;
x' represents X in Inputting vector output obtained by Embedding a token Embedding token;
then inputting a result LN (X ') obtained by carrying out layer normalization on the vector X' into a multi-head self-attention mechanism W-MSA based on a window to obtain a vector with extracted characteristic information, and adding the vector with X 'to obtain a vector X' with more collected characteristic information; x 'represents the input of X' into the window-based multi-headed autofocusing mechanism and the output resulting from the layer normalization operation;
continuously carrying out layer normalization on the vector X ', inputting the normalized LN (X ') into a local enhanced forward propagation network to obtain a vector with more extracted local characteristic information, and then adding the vector with X ' to obtain a vector X with more gathered local characteristic information out ，X out Represents the input of X "into the local enhanced forward propagation network and the resulting output of the layer normalization operation.
(3) Local enhanced forward propagation network (LeFF)
To enhance the ability of the image rendering model to capture local features, we add convolution to the forward propagation network, forming a local enhanced forward propagation network (LeFF), the specific design is shown in fig. 3 (c): firstly, an input sequence is changed into an image through a sequence changing module, the sequence is changed into the image, the image is convoluted by a convolution kernel of 1 × 1, then the image is activated through an activation function, then the image is convoluted by a convolution kernel of 3 × 3 and the convolution kernel of 1 × 1, then the image is activated through the activation function, and finally the image is changed into the sequence again to complete the local enhancement forward propagation network.
(4) Design of discriminator D
The essence of the discriminator is to determine whether the given data is "real", i.e. to determine whether the given data is real training data or false data generated by the color image generator G, as shown in fig. 3 (a), we first input real color images or false color images, and unwind them into small blocks (patch) by linear flattening, linearly flatten the blocks composed of convolution layers, then stack 4 MWin Transformer blocks having the same structure as G, and finally output true or false through the linear layers, thereby realizing the function of discrimination.
Example 1
In order to prove the effectiveness of the image coloring model, experiments are respectively carried out on an animal face image and a landscape image, and the image coloring model is compared with other current popular models: yoo et al (Yoo, S., bahng, H., chung, S., lee, J., chang, J., choo, J.: colouring with limited data: raw-shot colour assessed networks. In: proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.11283-11292 (2019)), su et al (Su, J. -W., chu, H., K., huang, J.: instance-aware image orientation. In: proceedings of the IEEE/F Conference and Pattern Recognition, pp.7968-7968)), and we compare the three indices: the Frechet increment distance score, the peak signal-to-noise ratio and the structural similarity are improved on the test of the two data sets, and the three indexes are improved by 0.003, 0.263 and 0.014 at the lowest. We find the more complex scenes the more realistic the details of model coloring, and we assume this is because a transform learns more about the distribution of large amounts of data than a CNN. In addition, we observed that the overall coloring effect of our model was smoother and more uniform than other methods, and without excessive color discontinuities, which demonstrates that the transformer can better capture the global information of the image.
Claims (7)
1. Image rendering method based on Transformer and generative confrontation network, characterized in that it is implemented according to the following steps,
step 1, constructing an image coloring model based on a generation countermeasure network, wherein the image coloring model comprises a color image generator and a discriminator; the color image generator is used for generating a color image, and the discriminator is used for judging whether the input image is a real color image or a false color image;
step 2, inputting the gray image into a color image generator of the image coloring model to generate a pseudo color image;
and 3, respectively updating parameters of the discriminator and the color image generator:
step 3.1: firstly, fixing parameters of a color image generator, inputting the false color image and the real color image corresponding to the gray image into an identifier in sequence, then calculating the loss between the real color image corresponding to the gray image and a label value of 1 according to a loss function, calculating the loss between the false color image generated by the gray image and a label value of 0 according to the loss function, and finally updating the parameters of the identifier by using a back propagation algorithm; wherein a label value of 1 represents a real image and a label value of 0 represents a generated pseudo-color image;
step 3.2: fixing parameters of the discriminator, calculating the loss between the generated pseudo color image and the label value of 1 according to a loss function, and finally updating the parameters of the color image generator by utilizing a back propagation algorithm;
step 3.3: continuously circulating the process of updating parameters of the discriminator and the color image generator in the steps 3.1 and 3.2 until the loss value is converged, and generating a false color image with good effect by the color image generator, namely obtaining an optimized image coloring model;
and 4, directly coloring the gray image by using the optimized image coloring model.
2. The method for rendering images based on Transformer and generation countermeasure network as claimed in claim 1, wherein in step 1, the color image generator comprises multiple MWin-Transformer modules, and the function of the MWin-Transformer module is to extract and reconstruct the features of the images, and output a 3-channel effective color image: the Mwin-transformer module consists of three core parts: a window-based multi-head self-attention mechanism, a layer normalization operation LN, and a local enhanced forward propagation network LeFF.
3. The Transformer-based and confrontational network-generating image rendering method of claim 2, wherein the color image generator generates a pseudo-color image as follows:
X′＝Embedded Tokens(X in )
X″＝W-MSA(LN(X))+X′
X out ＝LeFF(LN(X″))+X″
wherein X in Representing an input as a gray image or a false color image;
embedding Tokens denotes to assign X to in Converting into a vector;
x' represents X in Inputting the vector obtained by Embedding Tokens and outputting the vector;
then inputting a result LN (X ') obtained by carrying out layer normalization on the vector X' into a multi-head self-attention mechanism W-MSA based on a window to obtain a vector with extracted characteristic information, and adding the vector with X 'to obtain a vector X' with more collected characteristic information; x 'represents the input of X' into the output of the window-based multi-headed autofocusing mechanism and layer normalization operation;
continuously carrying out layer normalization on the vector X ', inputting the normalized LN (X ') into a local enhanced forward propagation network to obtain a vector with more extracted local characteristic information, and adding the vector with X ' to obtain a vector X with more gathered local characteristic information out ，X out Represents the input of X "into the locally enhanced forward propagation network LeFF and the resulting output of the layer normalization operation.
4. The transform-and-generative-confrontation-network-based image rendering method according to claim 3, wherein the layer normalization operation is calculated by:
wherein, the action object of LN layer is
X represents a vector, μ and δ represent the mean sum of each sampleThe variance of the measured values is calculated,
and
as affine learning parameters, d k Is the dimension of the concealment that is to be hidden,
indicating that the number is a k-dimensional vector.
5. The transform-based and generative countermeasure network-based image rendering method of claim 3, wherein the window-based multi-headed self-attention mechanism is as follows:
the method comprises the steps of dividing a pseudo-color image into a plurality of windows, and then performing self-attention calculation in the different windows, wherein the patch number in one window is far less than all small blocks in one picture, and the number of the windows is kept unchanged, so that the calculation complexity of the multi-head self-attention mechanism based on the windows and the size of the image are changed into a linear relation from a square relation, and the calculation complexity of a model is greatly reduced.
6. The transform-and-generate countermeasure-network-based image shading method of claim 2, wherein a convolution is added to the forward propagation network in the Mwin-transform module, thereby forming a local enhanced forward propagation network LeFF.
7. The Transformer-based and generative countermeasure network-based image rendering method of claim 1, wherein the loss function is:
wherein,
wherein, G * The sum of the loss functions is represented,
the presentation conditions are generated to counter network loss,
a weighting coefficient representing a Chardonnier loss, and λ represents the Chardonnier loss;
x represents an input gray image;
y represents a real color image corresponding to the input gray image;
log represents a base 2 logarithmic function;
ε represents a value of 10 -3 Constant coefficient of (d);
the absolute value is obtained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211247125.5A CN115546338A (en) | 2022-10-12 | 2022-10-12 | Image coloring method based on Transformer and generation countermeasure network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211247125.5A CN115546338A (en) | 2022-10-12 | 2022-10-12 | Image coloring method based on Transformer and generation countermeasure network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115546338A true CN115546338A (en) | 2022-12-30 |
Family
ID=84733928
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211247125.5A Pending CN115546338A (en) | 2022-10-12 | 2022-10-12 | Image coloring method based on Transformer and generation countermeasure network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115546338A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115908617A (en) * | 2023-01-09 | 2023-04-04 | 长春理工大学 | Infrared image colorizing method and system |
| CN116137043A (en) * | 2023-02-21 | 2023-05-19 | 长春理工大学 | Infrared image colorization method based on convolution and transfomer |
| CN116433788A (en) * | 2023-02-24 | 2023-07-14 | 北京科技大学 | Gray image coloring method and device based on self-attention and generation countermeasure network |
-
2022
- 2022-10-12 CN CN202211247125.5A patent/CN115546338A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115908617A (en) * | 2023-01-09 | 2023-04-04 | 长春理工大学 | Infrared image colorizing method and system |
| CN115908617B (en) * | 2023-01-09 | 2024-06-07 | 长春理工大学 | Infrared image colorization method and system |
| CN116137043A (en) * | 2023-02-21 | 2023-05-19 | 长春理工大学 | Infrared image colorization method based on convolution and transfomer |
| CN116433788A (en) * | 2023-02-24 | 2023-07-14 | 北京科技大学 | Gray image coloring method and device based on self-attention and generation countermeasure network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111145116B (en) | Sea surface rainy day image sample augmentation method based on generation of countermeasure network | |
| CN111798369B (en) | Face aging image synthesis method for generating confrontation network based on circulation condition | |
| CN115546338A (en) | Image coloring method based on Transformer and generation countermeasure network | |
| Wang et al. | A reinforcement learning paradigm of configuring visual enhancement for object detection in underwater scenes | |
| CN108961350B (en) | Wind painting migration method based on saliency matching | |
| CN109815893A (en) | The normalized method in colorized face images illumination domain of confrontation network is generated based on circulation | |
| CN112507793A (en) | Ultra-short-term photovoltaic power prediction method | |
| CN112489164B (en) | Image coloring method based on improved depth separable convolutional neural network | |
| CN113283444B (en) | Heterogeneous image migration method based on generation countermeasure network | |
| CN112002009B (en) | Unsupervised three-dimensional face reconstruction method based on generation of confrontation network | |
| Xiao et al. | Progressive data augmentation method for remote sensing ship image classification based on imaging simulation system and neural style transfer | |
| CN113989100B (en) | Infrared texture sample expansion method based on style generation countermeasure network | |
| CN111914617B (en) | Face attribute editing method based on balanced stack type generation type countermeasure network | |
| CN111401132A (en) | Pedestrian attribute identification method guided by high-level semantics under monitoring scene | |
| CN112270300A (en) | Method for converting human face sketch image into RGB image based on generating type confrontation network | |
| CN114399533B (en) | Single-target tracking method based on multi-level attention mechanism | |
| CN114581552A (en) | Gray level image colorizing method based on generation countermeasure network | |
| Zhang et al. | MSGAN: generative adversarial networks for image seasonal style transfer | |
| CN118230176B (en) | Wetland classification method and system based on multisource remote sensing and space-time depth integrated learning | |
| CN118172499B (en) | Building height inversion method based on resource third-order remote sensing image | |
| CN113888399B (en) | Face age synthesis method based on style fusion and domain selection structure | |
| CN116935043A (en) | Typical object remote sensing image generation method based on multitasking countermeasure network | |
| KumarSingh et al. | An Enhanced Image Colorization using Modified Generative Adversarial Networks with Pix2Pix Method | |
| Lee et al. | Design of CycleGAN model for SAR image colorization | |
| CN116137043A (en) | Infrared image colorization method based on convolution and transfomer |
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 |