SlideShare a Scribd company logo
1 of 19
Iranian Society of Machine Vision and Image Processing (ISMVIP)
Intro to Generative
Adversarial Networks (GANs)
Pegah Salehi
2020 - 17 - Dec
Email: pghsalehi@gmail.com
Generative Adversarial Networks (GANs)
[https://www.scopus.com/]18-1
Architecture
s
Applications
Main
Concept
Reference
Challenges
Introductio
n
2020
[https://thispersondoesnotexist.com/]
Iranian Society of Machine Vision and Image Processing (ISMVIP)
The GAN framework
Real Data
D
G
Fake Sample
JD
JG
𝐷(𝑥)
Real
or
FakeRandom Noise
18-2
[1]
Architecture
s
Applications
Main
Concept
Reference
Challenges
Introductio
n
Iranian Society of Machine Vision and Image Processing (ISMVIP)
The GAN framework
(1)
(2)
(3)
X 𝑑𝑎𝑡𝑎
D
Gz 𝐺(𝑧)
JD
JG
𝐷(𝑥)
Real
or
Fake
18-3
[1]
Architecture
s
Applications
Main
Concept
Reference
Challenges
Introductio
n
Iranian Society of Machine Vision and Image Processing (ISMVIP)
Training process
18-4
Architecture
s
Applications
Main
Concept
Reference
Challenges
Introductio
n
Iranian Society of Machine Vision and Image Processing (ISMVIP)
Challenges
• Non-convergence: the model parameters oscillate,
destabilize and never converge
• Mode collapse: the generator collapses which
produces limited varieties of samples
• Diminished gradient: the discriminator gets too
successful that the generator gradient vanishes and
learns nothing
18-5
Architecture
s
Applications
Main
Concept
Reference
Challenges
Introductio
n
Iranian Society of Machine Vision and Image Processing (ISMVIP)
Generate New Sample of Image Dataset
DCGA
N
BigGA
N
StyleGA
N
Characteristic
animation
18-6
Architecture
s
Applications
Main
Concept
Reference
Challenges
Introductio
n
Iranian Society of Machine Vision and Image Processing (ISMVIP)
semantic-image-to-photo translation
http://nvidia-research-mingyuliu.com/gaugan/
GauGA
N
Pix2Pix_H
D
18-7
Architecture
s
Applications
Main
Concept
Reference
Challenges
Introductio
n
Iranian Society of Machine Vision and Image Processing (ISMVIP)
https://www.youtube.com/watch?v=cQ54GDm1eL0&feature=youtu.be&t=32
Deepfake videos
18-8
Architecture
s
Applications
Main
Concept
Reference
Challenges
Introductio
n
Iranian Society of Machine Vision and Image Processing (ISMVIP)
Text-to-Image Translation
stack
GAN
18-9
Architecture
s
Applications
Main
Concept
Reference
Challenges
Introductio
n
Iranian Society of Machine Vision and Image Processing (ISMVIP)
Types of GAN Models
Architectural
Optimization
Objective Function
Optimization
Autoencoder
Conditional
Convolutional
DCGAN
CGAN, infoGAN,
ACGAN, SGAN
AAE, BiGAN, ALI,
AGE, VAE-GAN
Unrolled GAN, f-GAN, Mode-Regularized
GAN, Least-Square GAN, EBGAN,
WGAN, WGAN-GP, WGAN-LP
GenerativeAdversarialNetwork
(GAN)
18-10
[1]
Architecture
s
Applications
Main
Concept
Reference
Challenges
Introductio
n
Iranian Society of Machine Vision and Image Processing (ISMVIP)
32×32×128
4×4×1024
z
Decov 1
Decov 2
Decov 3
Decov 4
8×8×512
16×16×256
64×64×3
G(z)
100 Output
DCGAN
18-11
[1]
Architecture
s
Applications
Main
Concept
Reference
Challenges
Introductio
n
Iranian Society of Machine Vision and Image Processing (ISMVIP)
X
D
Gz G z c
Real
or
Fake
C (class)
cGAN
18-12
• Goal: Better control of the generation
• Idea: Add information about the generated sample (e.g., labels) to train the generator
[1]
Architecture
s
Applications
Main
Concept
Reference
Challenges
Introductio
n
Iranian Society of Machine Vision and Image Processing (ISMVIP)
Image-to-Image Translation
Pair Unpair
18-13
Architecture
s
Applications
Main
Concept
Reference
Challenges
Introductio
n
Iranian Society of Machine Vision and Image Processing (ISMVIP)
Pix2Pix
Fake Pair
Real
or
Fake
Input
Ground Truth (Reference)
U-net
PatchGAN
Discriminator
Generator
Generated
Tune G by computing L 1 distance
between output and ground truth.
Compute
Adversarial
Loss
Tune G
Tune D
Real Pair
18-14
Architecture
s
Applications
Main
Concept
Reference
Challenges
Introductio
n
Iranian Society of Machine Vision and Image Processing (ISMVIP)
[towardsdatascience.com]
cycleGAN
18-15
Architecture
s
Applications
Main
Concept
Reference
Challenges
Introductio
n
Iranian Society of Machine Vision and Image Processing (ISMVIP)
Introducing Education
Resources
18-16
Architecture
s
Applications
Main
Concept
Reference
Challenges
Introductio
n
Iranian Society of Machine Vision and Image Processing (ISMVIP)
References
1. Salehi, Pegah, Abdolah Chalechale, and Maryam
Taghizadeh. "Generative Adversarial Networks (GANs):
An Overview of Theoretical Model, Evaluation Metrics,
and Recent Developments." arXiv preprint
arXiv:2005.13178 (2020).
18-17
Architecture
s
Applications
Main
Concept
Reference
Challenges
Introductio
n
Iranian Society of Machine Vision and Image Processing (ISMVIP)
Thank you
for your attention...
18-18
Iranian Society of Machine Vision and Image Processing (ISMVIP)

More Related Content

Similar to Generative Adversarial Networks (GANs)

Computer Graphics Practical
Computer Graphics PracticalComputer Graphics Practical
Computer Graphics Practical
Neha Sharma
 
Finding connections among images using CycleGAN
Finding connections among images using CycleGANFinding connections among images using CycleGAN
Finding connections among images using CycleGAN
NAVER Engineering
 
Vladislav Kolbasin “Introduction to Generative Adversarial Networks (GANs)”
Vladislav Kolbasin “Introduction to Generative Adversarial Networks (GANs)”Vladislav Kolbasin “Introduction to Generative Adversarial Networks (GANs)”
Vladislav Kolbasin “Introduction to Generative Adversarial Networks (GANs)”
Lviv Startup Club
 

Similar to Generative Adversarial Networks (GANs) (20)

lec19.pdf
lec19.pdflec19.pdf
lec19.pdf
 
Computer Graphics Practical
Computer Graphics PracticalComputer Graphics Practical
Computer Graphics Practical
 
LSGAN - SIMPle(Simple Idea Meaningful Performance Level up)
LSGAN - SIMPle(Simple Idea Meaningful Performance Level up)LSGAN - SIMPle(Simple Idea Meaningful Performance Level up)
LSGAN - SIMPle(Simple Idea Meaningful Performance Level up)
 
CAR DAMAGE DETECTION USING DEEP LEARNING
CAR DAMAGE DETECTION USING DEEP LEARNINGCAR DAMAGE DETECTION USING DEEP LEARNING
CAR DAMAGE DETECTION USING DEEP LEARNING
 
Tracxn - Top Business Models - China Tech - Apr 2022
Tracxn - Top Business Models - China Tech  - Apr 2022Tracxn - Top Business Models - China Tech  - Apr 2022
Tracxn - Top Business Models - China Tech - Apr 2022
 
TEXT TO IMAGE GENERATION USING GAN
TEXT TO IMAGE GENERATION USING GANTEXT TO IMAGE GENERATION USING GAN
TEXT TO IMAGE GENERATION USING GAN
 
IRJET- Generating 3D Models Using 3D Generative Adversarial Network
IRJET- Generating 3D Models Using 3D Generative Adversarial NetworkIRJET- Generating 3D Models Using 3D Generative Adversarial Network
IRJET- Generating 3D Models Using 3D Generative Adversarial Network
 
EuroSciPy 2019 - GANs: Theory and Applications
EuroSciPy 2019 - GANs: Theory and ApplicationsEuroSciPy 2019 - GANs: Theory and Applications
EuroSciPy 2019 - GANs: Theory and Applications
 
[DLHacks 実装]Perceptual Adversarial Networks for Image-to-Image Transformation
[DLHacks 実装]Perceptual Adversarial Networks for Image-to-Image Transformation[DLHacks 実装]Perceptual Adversarial Networks for Image-to-Image Transformation
[DLHacks 実装]Perceptual Adversarial Networks for Image-to-Image Transformation
 
Tackling Open Images Challenge (2019)
Tackling Open Images Challenge (2019)Tackling Open Images Challenge (2019)
Tackling Open Images Challenge (2019)
 
Tracxn - Top Business Models - AR VR - Industry Applications - Apr 2022
Tracxn - Top Business Models - AR VR - Industry Applications - Apr 2022Tracxn - Top Business Models - AR VR - Industry Applications - Apr 2022
Tracxn - Top Business Models - AR VR - Industry Applications - Apr 2022
 
Finding connections among images using CycleGAN
Finding connections among images using CycleGANFinding connections among images using CycleGAN
Finding connections among images using CycleGAN
 
An Intelligent approach to Pic to Cartoon Conversion using White-box-cartooni...
An Intelligent approach to Pic to Cartoon Conversion using White-box-cartooni...An Intelligent approach to Pic to Cartoon Conversion using White-box-cartooni...
An Intelligent approach to Pic to Cartoon Conversion using White-box-cartooni...
 
Image super resolution using Generative Adversarial Network.
Image super resolution using Generative Adversarial Network.Image super resolution using Generative Adversarial Network.
Image super resolution using Generative Adversarial Network.
 
Tracxn - Top Business Models - AR VR Industry Applications Tech - Feb 2022
Tracxn  - Top Business Models - AR VR Industry Applications Tech - Feb 2022Tracxn  - Top Business Models - AR VR Industry Applications Tech - Feb 2022
Tracxn - Top Business Models - AR VR Industry Applications Tech - Feb 2022
 
IRJET- Transformation of Realistic Images and Videos into Cartoon Images and ...
IRJET- Transformation of Realistic Images and Videos into Cartoon Images and ...IRJET- Transformation of Realistic Images and Videos into Cartoon Images and ...
IRJET- Transformation of Realistic Images and Videos into Cartoon Images and ...
 
Vladislav Kolbasin “Introduction to Generative Adversarial Networks (GANs)”
Vladislav Kolbasin “Introduction to Generative Adversarial Networks (GANs)”Vladislav Kolbasin “Introduction to Generative Adversarial Networks (GANs)”
Vladislav Kolbasin “Introduction to Generative Adversarial Networks (GANs)”
 
IMAGE GENERATION WITH GANS-BASED TECHNIQUES: A SURVEY
IMAGE GENERATION WITH GANS-BASED TECHNIQUES: A SURVEYIMAGE GENERATION WITH GANS-BASED TECHNIQUES: A SURVEY
IMAGE GENERATION WITH GANS-BASED TECHNIQUES: A SURVEY
 
Image Generation with Gans-based Techniques: A Survey
Image Generation with Gans-based Techniques: A SurveyImage Generation with Gans-based Techniques: A Survey
Image Generation with Gans-based Techniques: A Survey
 
GAN - Theory and Applications
GAN - Theory and ApplicationsGAN - Theory and Applications
GAN - Theory and Applications
 

Recently uploaded

The title is not connected to what is inside
The title is not connected to what is insideThe title is not connected to what is inside
The title is not connected to what is inside
shinachiaurasa2
 
+971565801893>>SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHAB...
+971565801893>>SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHAB...+971565801893>>SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHAB...
+971565801893>>SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHAB...
Health
 
AI Mastery 201: Elevating Your Workflow with Advanced LLM Techniques
AI Mastery 201: Elevating Your Workflow with Advanced LLM TechniquesAI Mastery 201: Elevating Your Workflow with Advanced LLM Techniques
AI Mastery 201: Elevating Your Workflow with Advanced LLM Techniques
VictorSzoltysek
 

Recently uploaded (20)

Sector 18, Noida Call girls :8448380779 Model Escorts | 100% verified
Sector 18, Noida Call girls :8448380779 Model Escorts | 100% verifiedSector 18, Noida Call girls :8448380779 Model Escorts | 100% verified
Sector 18, Noida Call girls :8448380779 Model Escorts | 100% verified
 
Unveiling the Tech Salsa of LAMs with Janus in Real-Time Applications
Unveiling the Tech Salsa of LAMs with Janus in Real-Time ApplicationsUnveiling the Tech Salsa of LAMs with Janus in Real-Time Applications
Unveiling the Tech Salsa of LAMs with Janus in Real-Time Applications
 
A Secure and Reliable Document Management System is Essential.docx
A Secure and Reliable Document Management System is Essential.docxA Secure and Reliable Document Management System is Essential.docx
A Secure and Reliable Document Management System is Essential.docx
 
HR Software Buyers Guide in 2024 - HRSoftware.com
HR Software Buyers Guide in 2024 - HRSoftware.comHR Software Buyers Guide in 2024 - HRSoftware.com
HR Software Buyers Guide in 2024 - HRSoftware.com
 
Software Quality Assurance Interview Questions
Software Quality Assurance Interview QuestionsSoftware Quality Assurance Interview Questions
Software Quality Assurance Interview Questions
 
10 Trends Likely to Shape Enterprise Technology in 2024
10 Trends Likely to Shape Enterprise Technology in 202410 Trends Likely to Shape Enterprise Technology in 2024
10 Trends Likely to Shape Enterprise Technology in 2024
 
LEVEL 5 - SESSION 1 2023 (1).pptx - PDF 123456
LEVEL 5   - SESSION 1 2023 (1).pptx - PDF 123456LEVEL 5   - SESSION 1 2023 (1).pptx - PDF 123456
LEVEL 5 - SESSION 1 2023 (1).pptx - PDF 123456
 
The title is not connected to what is inside
The title is not connected to what is insideThe title is not connected to what is inside
The title is not connected to what is inside
 
The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...
The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...
The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...
 
Crypto Cloud Review - How To Earn Up To $500 Per DAY Of Bitcoin 100% On AutoP...
Crypto Cloud Review - How To Earn Up To $500 Per DAY Of Bitcoin 100% On AutoP...Crypto Cloud Review - How To Earn Up To $500 Per DAY Of Bitcoin 100% On AutoP...
Crypto Cloud Review - How To Earn Up To $500 Per DAY Of Bitcoin 100% On AutoP...
 
MarTech Trend 2024 Book : Marketing Technology Trends (2024 Edition) How Data...
MarTech Trend 2024 Book : Marketing Technology Trends (2024 Edition) How Data...MarTech Trend 2024 Book : Marketing Technology Trends (2024 Edition) How Data...
MarTech Trend 2024 Book : Marketing Technology Trends (2024 Edition) How Data...
 
%in Stilfontein+277-882-255-28 abortion pills for sale in Stilfontein
%in Stilfontein+277-882-255-28 abortion pills for sale in Stilfontein%in Stilfontein+277-882-255-28 abortion pills for sale in Stilfontein
%in Stilfontein+277-882-255-28 abortion pills for sale in Stilfontein
 
Direct Style Effect Systems - The Print[A] Example - A Comprehension Aid
Direct Style Effect Systems -The Print[A] Example- A Comprehension AidDirect Style Effect Systems -The Print[A] Example- A Comprehension Aid
Direct Style Effect Systems - The Print[A] Example - A Comprehension Aid
 
+971565801893>>SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHAB...
+971565801893>>SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHAB...+971565801893>>SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHAB...
+971565801893>>SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHAB...
 
Introducing Microsoft’s new Enterprise Work Management (EWM) Solution
Introducing Microsoft’s new Enterprise Work Management (EWM) SolutionIntroducing Microsoft’s new Enterprise Work Management (EWM) Solution
Introducing Microsoft’s new Enterprise Work Management (EWM) Solution
 
%in ivory park+277-882-255-28 abortion pills for sale in ivory park
%in ivory park+277-882-255-28 abortion pills for sale in ivory park %in ivory park+277-882-255-28 abortion pills for sale in ivory park
%in ivory park+277-882-255-28 abortion pills for sale in ivory park
 
BUS PASS MANGEMENT SYSTEM USING PHP.pptx
BUS PASS MANGEMENT SYSTEM USING PHP.pptxBUS PASS MANGEMENT SYSTEM USING PHP.pptx
BUS PASS MANGEMENT SYSTEM USING PHP.pptx
 
Optimizing AI for immediate response in Smart CCTV
Optimizing AI for immediate response in Smart CCTVOptimizing AI for immediate response in Smart CCTV
Optimizing AI for immediate response in Smart CCTV
 
AI Mastery 201: Elevating Your Workflow with Advanced LLM Techniques
AI Mastery 201: Elevating Your Workflow with Advanced LLM TechniquesAI Mastery 201: Elevating Your Workflow with Advanced LLM Techniques
AI Mastery 201: Elevating Your Workflow with Advanced LLM Techniques
 
ManageIQ - Sprint 236 Review - Slide Deck
ManageIQ - Sprint 236 Review - Slide DeckManageIQ - Sprint 236 Review - Slide Deck
ManageIQ - Sprint 236 Review - Slide Deck
 

Generative Adversarial Networks (GANs)