CN107871142A - A kind of empty convolution method based on depth convolution confrontation network model - Google Patents
A kind of empty convolution method based on depth convolution confrontation network model Download PDFInfo
- Publication number
- CN107871142A CN107871142A CN201711123735.3A CN201711123735A CN107871142A CN 107871142 A CN107871142 A CN 107871142A CN 201711123735 A CN201711123735 A CN 201711123735A CN 107871142 A CN107871142 A CN 107871142A
- Authority
- CN
- China
- Prior art keywords
- convolution
- mrow
- empty
- maker
- network model
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of empty convolution method based on depth convolution confrontation network model, belong to deep learning field of neural networks, described empty convolution method comprises the following steps:S1, construction are originally generated confrontation network model;S2, construction depth convolutional neural networks are as maker and arbiter;S3, initialization random noise, are inputted in maker;S4, convolution operation is carried out to image using empty convolution in neutral net;S5, loss function that empty convolution operation obtains input maker subsequently trained.The empty convolution method based on depth convolution confrontation network model of this method structure, change arbiter, maker receives the convolution mode after picture, arbiter, maker can be learnt with bigger scope to the feature of image, so as to improve the robustness of whole network training pattern.
Description
Technical field
The present invention relates to deep learning nerual network technique field, and in particular to one kind is based on depth convolution confrontation network mould
The empty convolution method of type.
Background technology
Production confrontation network (Generative Adversarial Network, abbreviation GAN) is by Goodfellow
In the deep learning framework that 2014 propose, it is based on the thought of " game theory ", construction maker (generator) and arbiter
(discriminator) two kinds of models, the former generates image by the Uniform noise or gaussian random noise for inputting (0,1), after
Person differentiates to the image of input, it is determined that being the image from data set or the image as caused by maker.
In traditional confrontation network model, operation that arbiter is carried out often by original convolution kernel, to generation
The feature of device generation image is parsed, and carries out a kind of study of new feature, generally requires to arrive a kind of old feature knot in study
It could be carried out after beam.In this case, the characteristic range that maker learns is smaller, and learning efficiency is relatively low.
The content of the invention
The invention aims to solve drawbacks described above of the prior art, there is provided one kind is based on depth convolution confrontation net
The empty convolution method of network model.
The purpose of the present invention can be reached by adopting the following technical scheme that:
A kind of empty convolution method described in empty convolution method based on depth convolution confrontation network model includes following
Step:
S1, construction are originally generated confrontation network model, and generating image by maker inputs to arbiter progress network instruction
Practice;
S2, construction depth convolutional neural networks are as maker and arbiter;
S3, initialization random noise, are inputted in maker;
S4, convolution operation is carried out to image using empty convolution in neutral net;
S5, loss function that empty convolution operation obtains input maker subsequently trained.
Further, described step S4 detailed processes are as follows:
S41, the multiple different numerical value of construction but size identical convolution kernel;
S42, using empty convolution convolution kernel is transformed, input network is trained.
Further, described step S5 detailed processes are as follows:
S51, the characteristics of image figure that will be obtained after empty convolution, input in arbiter and are differentiated;
S52, loss function that empty convolution operation obtains input maker subsequently trained.
Further, the expression formula of described loss function is:
Wherein, D (x) represents differentiation of the arbiter to image, and pr represents the distribution of data images, and pg represents generation image
Distribution, λ is hyper parameter,For gradient, E is the functional symbol for taking average.
The present invention is had the following advantages relative to prior art and effect:
Robustness:The present invention sets according to the operating process of empty convolution and constructs multiple empty convolution kernels, pass through convolution
Core inserts the mode of " 0 ", applies in the confrontation network model of maker and arbiter is served as with depth convolutional neural networks, more
It hurry up, more stably carry out network training.
Brief description of the drawings
Fig. 1 is the overall flow figure disclosed by the invention for being originally generated confrontation network and being trained by empty convolution;
Fig. 2 is the schematic diagram for being transformed into empty convolution kernel in the present invention to original convolution core.
Embodiment
To make the purpose, technical scheme and advantage of the embodiment of the present invention clearer, below in conjunction with the embodiment of the present invention
In accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is
Part of the embodiment of the present invention, rather than whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art
The every other embodiment obtained under the premise of creative work is not made, belongs to the scope of protection of the invention.
Embodiment
Present embodiment discloses a kind of empty convolution method based on depth convolution confrontation network model, specifically include following
Step:
Step S1, construction is originally generated confrontation network model, is inputted by maker generation image to arbiter and carries out net
Network training.
Step S2, construction depth convolutional neural networks are as maker and arbiter;
Different convolution kernels, it is embodied in difference, the difference of ranks number of matrix numerical value.
Multiple convolution kernels are constructed, during image is handled, different convolution kernels is meant in network training
Different characteristic of the process learning to generation image.
In the model of tradition confrontation network, the convolution kernel used in arbiter and maker is all fixed size and numerical value
Consistent, training effectiveness in this case is relatively low, and the characteristics of image scope learnt is relatively small.And at this
In invention, using empty convolution, the operation of " 0 " is interleave in being carried out to original convolution core, can be learnt so as to increase convolution kernel
The characteristic range arrived, further increase the efficiency of whole network study.
In actual applications, it should which according to the complexity of data images feature, the number of convolution kernel is set.
Step S3, random noise is initialized, is inputted in maker.
Step S4, convolution operation is carried out to image using empty convolution in neutral net.
Detailed process is as follows:
S41, the multiple different numerical value of construction but size identical convolution kernel;
S42, using the convolution kernel constructed, convolution is carried out to multiple images of maker generation respectively, it is more so as to obtain
Open characteristic pattern.
Step S5, the loss function input maker that empty convolution operation obtains subsequently is trained.Detailed process is such as
Under:
S51, by the characteristic pattern after convolution in step S4, input arbiter is differentiated;
S52, loss function that empty convolution operation obtains input maker subsequently trained;
S53, the average of all loss functions is inputted and continues to be trained into maker.
The effect of loss function is to weigh the ability that arbiter is judged generation image.The value of loss function is smaller, explanation
In current iteration, arbiter can have the generation image of preferable performance discrimination maker;Property that is on the contrary then illustrating arbiter
Can be poor.
The expression formula of loss function is:
Wherein, D (x) represents differentiation of the arbiter to image, and pr represents the distribution of data images, and pg represents generation image
Distribution, λ is hyper parameter,For gradient, E is the functional symbol for taking average.
In summary, present embodiment discloses a kind of empty convolution method based on depth convolution confrontation network model, phase
Than in traditional original confrontation network model, changing arbiter and receiving the mode learnt to characteristics of image after picture.
In the model of tradition confrontation network, arbiter is all fixed size with the convolution kernel used in maker and numerical value is consistent,
Training effectiveness in this case is relatively low, and the characteristics of image scope learnt is relatively small.And in the present invention,
Using empty convolution, the operation of " 0 " is interleave in being carried out to original convolution core, so as to increase the feature that convolution kernel can learn
Scope, further increase the efficiency of whole network study.
Above-described embodiment is the preferable embodiment of the present invention, but embodiments of the present invention are not by above-described embodiment
Limitation, other any Spirit Essences without departing from the present invention with made under principle change, modification, replacement, combine, simplification,
Equivalent substitute mode is should be, is included within protection scope of the present invention.
Claims (4)
- A kind of 1. empty convolution method based on depth convolution confrontation network model, it is characterised in that described empty convolution side Method comprises the following steps:S1, construction are originally generated confrontation network model, and generating image by maker inputs to arbiter progress network training;S2, construction depth convolutional neural networks are as maker and arbiter;S3, initialization random noise, are inputted in maker;S4, convolution operation is carried out to image using empty convolution in neutral net;S5, loss function that empty convolution operation obtains input maker subsequently trained.
- 2. a kind of empty convolution method based on depth convolution confrontation network model according to claim 1, its feature exist In described step S4 detailed processes are as follows:S41, the multiple different numerical value of construction but size identical convolution kernel;S42, using empty convolution convolution kernel is transformed, input network is trained.
- A kind of 3. depth convolution confrontation network model based on empty convolution according to claim 1, it is characterised in that institute The step S5 detailed processes stated are as follows:S51, the characteristics of image figure that will be obtained after empty convolution, input in arbiter and are differentiated;S52, loss function that empty convolution operation obtains input maker subsequently trained.
- A kind of 4. depth convolution confrontation network model based on empty convolution according to claim 3, it is characterised in that institute The expression formula for the loss function stated is:<mrow> <mi>L</mi> <mrow> <mo>(</mo> <mi>D</mi> <mo>)</mo> </mrow> <mo>=</mo> <msub> <mrow> <mo>-</mo> <mi>E</mi> </mrow> <mrow> <mi>x</mi> <mo>~</mo> <mi>pr</mi> </mrow> </msub> <mo>[</mo> <mi>D</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>)</mo> </mrow> <mo>]</mo> <mo>+</mo> <msub> <mi>E</mi> <mrow> <mi>x</mi> <mo>~</mo> <mi>pg</mi> </mrow> </msub> <mo>[</mo> <mi>D</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>)</mo> </mrow> <mo>]</mo> <mo>+</mo> <msub> <mi>&lambda;E</mi> <mrow> <mi>x</mi> <mo>~</mo> <mi>X</mi> </mrow> </msub> <msub> <mo>&dtri;</mo> <mi>x</mi> </msub> </mrow>Wherein, D (x) represents differentiation of the arbiter to image, and pr represents the distribution of data images, and pg represents point of generation image Cloth, λ are hyper parameter,For gradient, E is the functional symbol for taking average.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711123735.3A CN107871142A (en) | 2017-11-14 | 2017-11-14 | A kind of empty convolution method based on depth convolution confrontation network model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711123735.3A CN107871142A (en) | 2017-11-14 | 2017-11-14 | A kind of empty convolution method based on depth convolution confrontation network model |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107871142A true CN107871142A (en) | 2018-04-03 |
Family
ID=61753946
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711123735.3A Pending CN107871142A (en) | 2017-11-14 | 2017-11-14 | A kind of empty convolution method based on depth convolution confrontation network model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107871142A (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109214406A (en) * | 2018-05-16 | 2019-01-15 | 长沙理工大学 | Based on D-MobileNet neural network image classification method |
CN109345441A (en) * | 2018-10-19 | 2019-02-15 | 上海唯识律简信息科技有限公司 | A kind of de-watermarked method and system of image based on generation confrontation network |
CN109523538A (en) * | 2018-11-21 | 2019-03-26 | 上海七牛信息技术有限公司 | A kind of people counting method and system based on generation confrontation neural network |
CN109615059A (en) * | 2018-11-06 | 2019-04-12 | 海南大学 | Edge filling and filter dilation operation method and system in a kind of convolutional neural networks |
CN109741328A (en) * | 2019-02-02 | 2019-05-10 | 东北大学 | A kind of automobile apparent mass detection method based on production confrontation network |
CN110136731A (en) * | 2019-05-13 | 2019-08-16 | 天津大学 | Empty cause and effect convolution generates the confrontation blind Enhancement Method of network end-to-end bone conduction voice |
CN110363210A (en) * | 2018-04-10 | 2019-10-22 | 腾讯科技(深圳)有限公司 | A kind of training method and server of image, semantic parted pattern |
CN110610184A (en) * | 2018-06-15 | 2019-12-24 | 阿里巴巴集团控股有限公司 | Method, device and equipment for detecting salient object of image |
CN109784149B (en) * | 2018-12-06 | 2021-08-20 | 苏州飞搜科技有限公司 | Method and system for detecting key points of human skeleton |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106997380A (en) * | 2017-03-21 | 2017-08-01 | 北京工业大学 | Imaging spectrum safe retrieving method based on DCGAN depth networks |
CN107103590A (en) * | 2017-03-22 | 2017-08-29 | 华南理工大学 | A kind of image for resisting generation network based on depth convolution reflects minimizing technology |
CN107293289A (en) * | 2017-06-13 | 2017-10-24 | 南京医科大学 | A kind of speech production method that confrontation network is generated based on depth convolution |
-
2017
- 2017-11-14 CN CN201711123735.3A patent/CN107871142A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106997380A (en) * | 2017-03-21 | 2017-08-01 | 北京工业大学 | Imaging spectrum safe retrieving method based on DCGAN depth networks |
CN107103590A (en) * | 2017-03-22 | 2017-08-29 | 华南理工大学 | A kind of image for resisting generation network based on depth convolution reflects minimizing technology |
CN107293289A (en) * | 2017-06-13 | 2017-10-24 | 南京医科大学 | A kind of speech production method that confrontation network is generated based on depth convolution |
Non-Patent Citations (4)
Title |
---|
ALEC RADFORD ET AL.: "UNSUPERVISED REPRESENTATION LEARNING WITH DEEP CONVOLUTIONAL GENERATIVE ADVERSARIAL NETWORKS", 《MARCHINE LEARNING》 * |
ANDREW BROK ET AL.: "NEURAL PHOTO EDITING WITH INTROSPECTIVE ADVERSARIAL NETWORKS", 《MACHINE LEARNING》 * |
ISHAAN GULRAJANI ET AL.: "Improved Training of Wasserstein GANs", 《MACHINE LEARNING》 * |
徐一峰: "生成对抗网络理论模型和应用综述", 《金华职业技术学院学报》 * |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110363210A (en) * | 2018-04-10 | 2019-10-22 | 腾讯科技(深圳)有限公司 | A kind of training method and server of image, semantic parted pattern |
CN110363210B (en) * | 2018-04-10 | 2023-05-05 | 腾讯科技(深圳)有限公司 | Training method and server for image semantic segmentation model |
CN109214406A (en) * | 2018-05-16 | 2019-01-15 | 长沙理工大学 | Based on D-MobileNet neural network image classification method |
CN110610184A (en) * | 2018-06-15 | 2019-12-24 | 阿里巴巴集团控股有限公司 | Method, device and equipment for detecting salient object of image |
CN110610184B (en) * | 2018-06-15 | 2023-05-12 | 阿里巴巴集团控股有限公司 | Method, device and equipment for detecting salient targets of images |
CN109345441A (en) * | 2018-10-19 | 2019-02-15 | 上海唯识律简信息科技有限公司 | A kind of de-watermarked method and system of image based on generation confrontation network |
CN109615059A (en) * | 2018-11-06 | 2019-04-12 | 海南大学 | Edge filling and filter dilation operation method and system in a kind of convolutional neural networks |
CN109523538A (en) * | 2018-11-21 | 2019-03-26 | 上海七牛信息技术有限公司 | A kind of people counting method and system based on generation confrontation neural network |
CN109784149B (en) * | 2018-12-06 | 2021-08-20 | 苏州飞搜科技有限公司 | Method and system for detecting key points of human skeleton |
CN109741328A (en) * | 2019-02-02 | 2019-05-10 | 东北大学 | A kind of automobile apparent mass detection method based on production confrontation network |
CN109741328B (en) * | 2019-02-02 | 2023-04-14 | 东北大学 | Automobile apparent mass detection method based on generative confrontation network |
CN110136731A (en) * | 2019-05-13 | 2019-08-16 | 天津大学 | Empty cause and effect convolution generates the confrontation blind Enhancement Method of network end-to-end bone conduction voice |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107871142A (en) | A kind of empty convolution method based on depth convolution confrontation network model | |
CN107862377A (en) | A kind of packet convolution method that confrontation network model is generated based on text image | |
CN107590518A (en) | A kind of confrontation network training method of multiple features study | |
CN107886169A (en) | A kind of multiple dimensioned convolution kernel method that confrontation network model is generated based on text image | |
CN107886162A (en) | A kind of deformable convolution kernel method based on WGAN models | |
CN107590531A (en) | A kind of WGAN methods based on text generation | |
CN107944546A (en) | It is a kind of based on be originally generated confrontation network model residual error network method | |
CN107563493A (en) | A kind of confrontation network algorithm of more maker convolution composographs | |
CN107767413A (en) | A kind of image depth estimation method based on convolutional neural networks | |
CN107992944A (en) | It is a kind of based on be originally generated confrontation network model multiple dimensioned convolution method | |
CN108182657A (en) | A kind of face-image conversion method that confrontation network is generated based on cycle | |
CN106683067A (en) | Deep learning super-resolution reconstruction method based on residual sub-images | |
CN109389556A (en) | The multiple dimensioned empty convolutional neural networks ultra-resolution ratio reconstructing method of one kind and device | |
CN107705242A (en) | A kind of image stylization moving method of combination deep learning and depth perception | |
CN110222722A (en) | Interactive image stylization processing method, calculates equipment and storage medium at system | |
CN109934826A (en) | A kind of characteristics of image dividing method based on figure convolutional network | |
CN108009568A (en) | A kind of pedestrian detection method based on WGAN models | |
CN106339984B (en) | Distributed image ultra-resolution method based on K mean value driving convolutional neural networks | |
CN109360159A (en) | A kind of image completion method based on generation confrontation network model | |
CN106022392A (en) | Deep neural network sample automatic accepting and rejecting training method | |
CN107943750A (en) | A kind of decomposition convolution method based on WGAN models | |
CN106776540A (en) | A kind of liberalization document creation method | |
CN108765512A (en) | A kind of confrontation image generating method based on multi-layer feature | |
CN108805151A (en) | A kind of image classification method based on depth similitude network | |
CN108111860A (en) | Video sequence lost frames prediction restoration methods based on depth residual error network |
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: 20180403 |