SlideShare a Scribd company logo
1 of 25
visionNoob
(Jaewon Lee)
PR-122
CAN: Creative Adversarial Networks,
Generating "Art" by Learning About Styles and Deviating from
Style NormsAhmed Elgammal, Bingchen Liu, Mohamed Elhoseiny, Marian Mazzone, (2017)
1
https://arxiv.org/abs/1706.07068
2
∗This paper is an extended version of a paper published on the eighth International Conference on
Computational Creativity (ICCC), held in Atlanta, GA, June 20th-June 22nd, 2017.
The Art and Artificial Intelligence Laboratory at Rutgers: https://sites.google.com/site/digihumanlab/home
3
Figure 1: Example of images generated by CAN. The generated images vary from simple abstract ones to
complex textures and compositions. (256 x 256)
4
Figure 1: Example of images generated by CAN. The generated images vary from simple abstract ones to
complex textures and compositions. (256 x 256)
5
If we teach the machine about art and art styles and
force it to generate novel images that do not follow
established styles,
what would it generate?
6
Would it generate something
that is aesthetically appealing to humans?
CAN : Top ranked by human subjects 1
7
Would that be considered “art”?
CAN : Top ranked by human subjects 2
t
8
Art Generating Agent
The agent tries to generate that is novel, but not too novel.
https://www.psychologytoday.com/intl/blog/in-the-brain-the-beholder/201405/its-acquired-taste-how-knowledge-drives-aesthetics?amp
https://news.artnet.com/market/google-inceptionism-art-sells-big-439352
too novel
novel
deepdream
not
nobel
“dazzling, druggy, and creepy”
9
Background
Wundt curve
Novelty
Surprisingness
Complexity
Ambiguity
Puzzlingness
Computational Curiosity, Qiong Wu Published 2015 in ArXiv
t
10
Art Generating Agent
The agent tries to generate that is novel, but not too novel.
https://www.psychologytoday.com/intl/blog/in-the-brain-the-beholder/201405/its-acquired-taste-how-knowledge-drives-aesthetics?amp
https://news.artnet.com/market/google-inceptionism-art-sells-big-439352
“dazzling, druggy, and creepy”
too novel
novel
deepdream
not
nobel
11
Dataset
WikiArt Dataset (2015)
81,449 paintings from 1,119 artists (from 15th~21st century) with style label
https://www.wikiart.org/
12
GAN is Emulative and not Creative
Recap - Vanilla GAN
min
𝐺
max
𝐷
𝑉 𝐷, 𝐺 = 𝔼 𝑥~𝑝 𝑑𝑎𝑡𝑎
[log 𝐷 𝑥 ] + 𝔼 𝑧~𝑝 𝑧
[log(1 − 𝐷 𝐺(𝑧 ))]
https://blog.openai.com/generative-models/
13
min
𝐺
max
𝐷
𝑉 𝐷, 𝐺 = 𝔼 𝑥~𝑝 𝑑𝑎𝑡𝑎
[log 𝐷 𝑥 ] + 𝔼 𝑧~𝑝 𝑧
[log(1 − 𝐷 𝐺(𝑧 ))]
min
𝐺
max
𝐷
𝑉 𝐷, 𝐺 =
𝔼 𝑥, 𝑐~𝑝 𝑑𝑎𝑡𝑎
[log 𝐷𝑟 𝑥 + log 𝐷𝑐 𝑐 = 𝑐|𝑥 ] +
𝔼 𝑧~𝑝 𝑧
[log(1 − 𝐷𝑟 𝐺(𝑧 )) −
𝑘=1
𝐾
1
𝐾
log(𝐷𝑐 𝑐 𝑘 𝐺 𝑧 + (1 −
1
𝑘
) log(1 − 𝐷𝑐(𝑐 𝑘|𝐺 𝑧 ) ]
recap - Vanilla GAN loss
𝑐 : style label from the the data distribution 𝑝 𝑑𝑎𝑡𝑎
K : # of style classes
𝐷𝑟 〮 : discriminate between real art and generated images
𝐷𝑐(〮) : discriminate between different style categories and estimate posterior (i.e., 𝐷𝑐 𝑐 𝑘 〮 = 𝑝(𝑐 𝑘|〮))
style classification
style ambiguity
maximizes the entropy of p(c|x)
CAN : Loss function
14
CAN : Loss function
Training Discriminator
Training Generator
max
𝐷
𝑉 𝐷, 𝐺 = 𝔼 𝑥, 𝑐~𝑝 𝑑𝑎𝑡𝑎
[log 𝐷𝑟 𝑥 + log 𝐷𝑐 𝑐 = 𝑐|𝑥 ] + # For Training data
𝔼 𝑧~𝑝 𝑧
[log(1 − 𝐷 𝐺(𝑧 ))] # For Generated data
𝔼 𝑧~𝑝 𝑧
[log(1 − 𝐷 𝐺(𝑧 )) −
𝑘=1
𝐾
1
𝐾
log(𝐷𝑐 𝑐 𝑘 𝐺 𝑧 + (1 −
1
𝑘
) log(1 − 𝐷𝑐(𝑐 𝑘|𝐺 𝑧 ) ]
min
𝐺
𝑉 𝐷, 𝐺 =
style classification
style ambiguity
15
16
Experiments
CAN implementation : https://github.com/mlberkeley/Creative-Adversarial-Networks
Experiments
17
Basically…
Assessing the creativity of artifacts generated
by the machine is an open and hard question.
Dataset : WikiArt dataset
Baseline 1 : original DCGAN (64 x 64)
Baseline 2 : original DCGAN (256 x 256)
Baseline 3 : CAN (w.o. style ambiguity) (256 x 256)
Proposed : CAN (256 x 256)
18
We also do not see any recognizable figures.
Many of the images seems abstract.
Is that simply because it fails to emulate the art distribution or
Is it because it tried to generate novel images?
Is it at all creative?
Figure 1: Example of images generated by CAN. The generated images vary from simple abstract ones to
complex textures and compositions. (256 x 256)
19
Qualitative Result
DCGAN 64 x 64 DCGAN 256 x 256 CAN 256 x 256
(w.o. style ambiguity)
CAN 256 x 256
20
Quantitative Result
Experiment 1 (Amazone Mturk)
test the ability of the system to generate art that human users
could not distinguish from top creative art that is being generated by artists today
Q1: Do you think the work is created by an artist or generated by a computer?
The user has to choose one of two answers: artist or computer.
Q2: The user asked to rate how they like the image
in a scale 1 (extremely dislike) to 5 (extremely like).
21
Figure 8: Art Basel Set: a collection of 25
paintings selected from Art Basel 2017 art fair.
Shamir, Lior, Jenny Nissel, and Ellen Winner. "Distinguishing between abstract art by artists vs. children and animals: Comparison
between human and machine perception." ACM Transactions on Applied Perception (TAP) 13.3 (2016): 17
22
Quantitative Result
Experiment 2 (Amazone MTurk)
Q1 How do you like this image: 1-extremely dislike ~ 5-extremely like.
Q2 Rate the novelty of the image: 1-extremely not novel, ~ 5-extremely novel.
Q3 Do you find the image surprising: 1-extremely not surprising ~ 5-extremely surprising.
Q4 Rate the ambiguity of the image. I find this image: 1-extremely not ambiguous ~ 5-extremely ambiguous.
Q5 Rate the complexity of the image. I find this image: 1-extremely simple ~ 5-extremely complex
Q6 Do you think the image is created by an artist or generated by computer?
23
Quantitative Result
Experiment 3
judge aspects related to whether the images generated by CAN can be considered art
Q1: As I interact with this painting, I start to see the artist’s intentionality:
it looks like it was composed very intentionally.
Q2: As I interact with this painting, I start to see a structure emerging.
Q3: Communication: As I interact with this painting, I feel that it is communicating with me.
Q4: Inspiration: As I interact with this painting, I feel inspired and elevated.
24
Quantitative Result
Experiment 4 CAN vs sc-CAN (sophisticated art-educated subjects)
evaluate the effect of adding the style ambiguity loss to the CAN model
Q1 Which image do you think is more novel?
Q2 Which image do you think is more aesthetically appealing?
Result:
CAN images are more novel than sc-CAN : 59.47%
CAN images are more aesthetically appealing than sc-CAN : 60%
Q&A
25
Figure 5: Example of images generated by CAN. Top: Images ranked high in “likeness”
according to human subjects. Bottom: Images ranked the lowest by human subjects.

More Related Content

Similar to PR-122: Can-Creative Adversarial Networks

A Learned Representation for Artistic Style
A Learned Representation for Artistic StyleA Learned Representation for Artistic Style
A Learned Representation for Artistic StyleMayank Agarwal
 
Gan seminar
Gan seminarGan seminar
Gan seminarSan Kim
 
Steam presentation deux 3 d prints from photographs
Steam presentation deux  3 d prints from photographsSteam presentation deux  3 d prints from photographs
Steam presentation deux 3 d prints from photographsScott Eastellerson
 
Machine Intelligence.html
Machine Intelligence.htmlMachine Intelligence.html
Machine Intelligence.htmlJohnChan191
 
Pixelor presentation slides for SIGGRAPH Asia 2020
Pixelor presentation slides for SIGGRAPH Asia 2020Pixelor presentation slides for SIGGRAPH Asia 2020
Pixelor presentation slides for SIGGRAPH Asia 2020Ayan Das
 
Vlatko Ceric: My Art An Overview
Vlatko Ceric: My Art An OverviewVlatko Ceric: My Art An Overview
Vlatko Ceric: My Art An OverviewVlatko Ceric
 
June8 study neural_style_transfer_olivia_seji_oh
June8 study neural_style_transfer_olivia_seji_ohJune8 study neural_style_transfer_olivia_seji_oh
June8 study neural_style_transfer_olivia_seji_ohSeji OH
 
Creativity through deep learning
Creativity through deep learningCreativity through deep learning
Creativity through deep learningAkin Osman Kazakci
 
Deep learning in Computer Vision
Deep learning in Computer VisionDeep learning in Computer Vision
Deep learning in Computer VisionDavid Dao
 
Implementing Neural Style Transfer
Implementing Neural Style Transfer Implementing Neural Style Transfer
Implementing Neural Style Transfer Tahsin Mayeesha
 
2021_jiayuhe_portfolio.pdf
2021_jiayuhe_portfolio.pdf2021_jiayuhe_portfolio.pdf
2021_jiayuhe_portfolio.pdfJIAYU HE
 
Artist Assistant AI(AAA)
Artist Assistant AI(AAA)Artist Assistant AI(AAA)
Artist Assistant AI(AAA)Gunhee Lee
 
Deep Residual Hashing Neural Network for Image Retrieval
Deep Residual Hashing Neural Network for Image RetrievalDeep Residual Hashing Neural Network for Image Retrieval
Deep Residual Hashing Neural Network for Image RetrievalEdwin Efraín Jiménez Lepe
 
Rutgers Invited Talk: Creative Expression to Motivate Interest in Computing
Rutgers Invited Talk: Creative Expression to Motivate Interest in ComputingRutgers Invited Talk: Creative Expression to Motivate Interest in Computing
Rutgers Invited Talk: Creative Expression to Motivate Interest in ComputingMark Guzdial
 
Understanding Basics of Machine Learning
Understanding Basics of Machine LearningUnderstanding Basics of Machine Learning
Understanding Basics of Machine LearningPranav Ainavolu
 
The idea of projectour project is about creating a intell.docx
The idea of projectour project is about creating a intell.docxThe idea of projectour project is about creating a intell.docx
The idea of projectour project is about creating a intell.docxcherry686017
 
Faire de la reconnaissance d'images avec le Deep Learning - Cristina & Pierre...
Faire de la reconnaissance d'images avec le Deep Learning - Cristina & Pierre...Faire de la reconnaissance d'images avec le Deep Learning - Cristina & Pierre...
Faire de la reconnaissance d'images avec le Deep Learning - Cristina & Pierre...Jedha Bootcamp
 
PyDresden 20170824 - Deep Learning for Computer Vision
PyDresden 20170824 - Deep Learning for Computer VisionPyDresden 20170824 - Deep Learning for Computer Vision
PyDresden 20170824 - Deep Learning for Computer VisionAlex Conway
 

Similar to PR-122: Can-Creative Adversarial Networks (20)

Image generative modeling for design inspiration and image editing by Camille...
Image generative modeling for design inspiration and image editing by Camille...Image generative modeling for design inspiration and image editing by Camille...
Image generative modeling for design inspiration and image editing by Camille...
 
A Learned Representation for Artistic Style
A Learned Representation for Artistic StyleA Learned Representation for Artistic Style
A Learned Representation for Artistic Style
 
Gan seminar
Gan seminarGan seminar
Gan seminar
 
Steam presentation deux 3 d prints from photographs
Steam presentation deux  3 d prints from photographsSteam presentation deux  3 d prints from photographs
Steam presentation deux 3 d prints from photographs
 
Machine Intelligence.html
Machine Intelligence.htmlMachine Intelligence.html
Machine Intelligence.html
 
Pixelor presentation slides for SIGGRAPH Asia 2020
Pixelor presentation slides for SIGGRAPH Asia 2020Pixelor presentation slides for SIGGRAPH Asia 2020
Pixelor presentation slides for SIGGRAPH Asia 2020
 
Vlatko Ceric: My Art An Overview
Vlatko Ceric: My Art An OverviewVlatko Ceric: My Art An Overview
Vlatko Ceric: My Art An Overview
 
June8 study neural_style_transfer_olivia_seji_oh
June8 study neural_style_transfer_olivia_seji_ohJune8 study neural_style_transfer_olivia_seji_oh
June8 study neural_style_transfer_olivia_seji_oh
 
Creativity through deep learning
Creativity through deep learningCreativity through deep learning
Creativity through deep learning
 
Deep learning in Computer Vision
Deep learning in Computer VisionDeep learning in Computer Vision
Deep learning in Computer Vision
 
Implementing Neural Style Transfer
Implementing Neural Style Transfer Implementing Neural Style Transfer
Implementing Neural Style Transfer
 
2021_jiayuhe_portfolio.pdf
2021_jiayuhe_portfolio.pdf2021_jiayuhe_portfolio.pdf
2021_jiayuhe_portfolio.pdf
 
Artist Assistant AI(AAA)
Artist Assistant AI(AAA)Artist Assistant AI(AAA)
Artist Assistant AI(AAA)
 
Deep Residual Hashing Neural Network for Image Retrieval
Deep Residual Hashing Neural Network for Image RetrievalDeep Residual Hashing Neural Network for Image Retrieval
Deep Residual Hashing Neural Network for Image Retrieval
 
Rutgers Invited Talk: Creative Expression to Motivate Interest in Computing
Rutgers Invited Talk: Creative Expression to Motivate Interest in ComputingRutgers Invited Talk: Creative Expression to Motivate Interest in Computing
Rutgers Invited Talk: Creative Expression to Motivate Interest in Computing
 
CBIR_white.ppt
CBIR_white.pptCBIR_white.ppt
CBIR_white.ppt
 
Understanding Basics of Machine Learning
Understanding Basics of Machine LearningUnderstanding Basics of Machine Learning
Understanding Basics of Machine Learning
 
The idea of projectour project is about creating a intell.docx
The idea of projectour project is about creating a intell.docxThe idea of projectour project is about creating a intell.docx
The idea of projectour project is about creating a intell.docx
 
Faire de la reconnaissance d'images avec le Deep Learning - Cristina & Pierre...
Faire de la reconnaissance d'images avec le Deep Learning - Cristina & Pierre...Faire de la reconnaissance d'images avec le Deep Learning - Cristina & Pierre...
Faire de la reconnaissance d'images avec le Deep Learning - Cristina & Pierre...
 
PyDresden 20170824 - Deep Learning for Computer Vision
PyDresden 20170824 - Deep Learning for Computer VisionPyDresden 20170824 - Deep Learning for Computer Vision
PyDresden 20170824 - Deep Learning for Computer Vision
 

More from jaewon lee

PR-185: RetinaFace: Single-stage Dense Face Localisation in the Wild
PR-185: RetinaFace: Single-stage Dense Face Localisation in the WildPR-185: RetinaFace: Single-stage Dense Face Localisation in the Wild
PR-185: RetinaFace: Single-stage Dense Face Localisation in the Wildjaewon lee
 
PR-199: SNIPER:Efficient Multi Scale Training
PR-199: SNIPER:Efficient Multi Scale TrainingPR-199: SNIPER:Efficient Multi Scale Training
PR-199: SNIPER:Efficient Multi Scale Trainingjaewon lee
 
PR-146: CornerNet detecting objects as paired keypoints
PR-146: CornerNet detecting objects as paired keypointsPR-146: CornerNet detecting objects as paired keypoints
PR-146: CornerNet detecting objects as paired keypointsjaewon lee
 
PR 171: Large margin softmax loss for Convolutional Neural Networks
PR 171: Large margin softmax loss for Convolutional Neural NetworksPR 171: Large margin softmax loss for Convolutional Neural Networks
PR 171: Large margin softmax loss for Convolutional Neural Networksjaewon lee
 
PR157: Best of both worlds: human-machine collaboration for object annotation
PR157: Best of both worlds: human-machine collaboration for object annotationPR157: Best of both worlds: human-machine collaboration for object annotation
PR157: Best of both worlds: human-machine collaboration for object annotationjaewon lee
 
Pytorch kr devcon
Pytorch kr devconPytorch kr devcon
Pytorch kr devconjaewon lee
 
PR-134 How Does Batch Normalization Help Optimization?
PR-134 How Does Batch Normalization Help Optimization?PR-134 How Does Batch Normalization Help Optimization?
PR-134 How Does Batch Normalization Help Optimization?jaewon lee
 
PR-110: An Analysis of Scale Invariance in Object Detection – SNIP
PR-110: An Analysis of Scale Invariance in Object Detection – SNIPPR-110: An Analysis of Scale Invariance in Object Detection – SNIP
PR-110: An Analysis of Scale Invariance in Object Detection – SNIPjaewon lee
 

More from jaewon lee (9)

PR-185: RetinaFace: Single-stage Dense Face Localisation in the Wild
PR-185: RetinaFace: Single-stage Dense Face Localisation in the WildPR-185: RetinaFace: Single-stage Dense Face Localisation in the Wild
PR-185: RetinaFace: Single-stage Dense Face Localisation in the Wild
 
PR-199: SNIPER:Efficient Multi Scale Training
PR-199: SNIPER:Efficient Multi Scale TrainingPR-199: SNIPER:Efficient Multi Scale Training
PR-199: SNIPER:Efficient Multi Scale Training
 
PR-146: CornerNet detecting objects as paired keypoints
PR-146: CornerNet detecting objects as paired keypointsPR-146: CornerNet detecting objects as paired keypoints
PR-146: CornerNet detecting objects as paired keypoints
 
PR 171: Large margin softmax loss for Convolutional Neural Networks
PR 171: Large margin softmax loss for Convolutional Neural NetworksPR 171: Large margin softmax loss for Convolutional Neural Networks
PR 171: Large margin softmax loss for Convolutional Neural Networks
 
PR157: Best of both worlds: human-machine collaboration for object annotation
PR157: Best of both worlds: human-machine collaboration for object annotationPR157: Best of both worlds: human-machine collaboration for object annotation
PR157: Best of both worlds: human-machine collaboration for object annotation
 
Rgb data
Rgb dataRgb data
Rgb data
 
Pytorch kr devcon
Pytorch kr devconPytorch kr devcon
Pytorch kr devcon
 
PR-134 How Does Batch Normalization Help Optimization?
PR-134 How Does Batch Normalization Help Optimization?PR-134 How Does Batch Normalization Help Optimization?
PR-134 How Does Batch Normalization Help Optimization?
 
PR-110: An Analysis of Scale Invariance in Object Detection – SNIP
PR-110: An Analysis of Scale Invariance in Object Detection – SNIPPR-110: An Analysis of Scale Invariance in Object Detection – SNIP
PR-110: An Analysis of Scale Invariance in Object Detection – SNIP
 

Recently uploaded

Industrialised data - the key to AI success.pdf
Industrialised data - the key to AI success.pdfIndustrialised data - the key to AI success.pdf
Industrialised data - the key to AI success.pdfLars Albertsson
 
B2 Creative Industry Response Evaluation.docx
B2 Creative Industry Response Evaluation.docxB2 Creative Industry Response Evaluation.docx
B2 Creative Industry Response Evaluation.docxStephen266013
 
BigBuy dropshipping via API with DroFx.pptx
BigBuy dropshipping via API with DroFx.pptxBigBuy dropshipping via API with DroFx.pptx
BigBuy dropshipping via API with DroFx.pptxolyaivanovalion
 
Carero dropshipping via API with DroFx.pptx
Carero dropshipping via API with DroFx.pptxCarero dropshipping via API with DroFx.pptx
Carero dropshipping via API with DroFx.pptxolyaivanovalion
 
Generative AI on Enterprise Cloud with NiFi and Milvus
Generative AI on Enterprise Cloud with NiFi and MilvusGenerative AI on Enterprise Cloud with NiFi and Milvus
Generative AI on Enterprise Cloud with NiFi and MilvusTimothy Spann
 
Low Rate Call Girls Bhilai Anika 8250192130 Independent Escort Service Bhilai
Low Rate Call Girls Bhilai Anika 8250192130 Independent Escort Service BhilaiLow Rate Call Girls Bhilai Anika 8250192130 Independent Escort Service Bhilai
Low Rate Call Girls Bhilai Anika 8250192130 Independent Escort Service BhilaiSuhani Kapoor
 
04242024_CCC TUG_Joins and Relationships
04242024_CCC TUG_Joins and Relationships04242024_CCC TUG_Joins and Relationships
04242024_CCC TUG_Joins and Relationshipsccctableauusergroup
 
VIP Call Girls in Amravati Aarohi 8250192130 Independent Escort Service Amravati
VIP Call Girls in Amravati Aarohi 8250192130 Independent Escort Service AmravatiVIP Call Girls in Amravati Aarohi 8250192130 Independent Escort Service Amravati
VIP Call Girls in Amravati Aarohi 8250192130 Independent Escort Service AmravatiSuhani Kapoor
 
Midocean dropshipping via API with DroFx
Midocean dropshipping via API with DroFxMidocean dropshipping via API with DroFx
Midocean dropshipping via API with DroFxolyaivanovalion
 
RA-11058_IRR-COMPRESS Do 198 series of 1998
RA-11058_IRR-COMPRESS Do 198 series of 1998RA-11058_IRR-COMPRESS Do 198 series of 1998
RA-11058_IRR-COMPRESS Do 198 series of 1998YohFuh
 
Ravak dropshipping via API with DroFx.pptx
Ravak dropshipping via API with DroFx.pptxRavak dropshipping via API with DroFx.pptx
Ravak dropshipping via API with DroFx.pptxolyaivanovalion
 
Call me @ 9892124323 Cheap Rate Call Girls in Vashi with Real Photo 100% Secure
Call me @ 9892124323  Cheap Rate Call Girls in Vashi with Real Photo 100% SecureCall me @ 9892124323  Cheap Rate Call Girls in Vashi with Real Photo 100% Secure
Call me @ 9892124323 Cheap Rate Call Girls in Vashi with Real Photo 100% SecurePooja Nehwal
 
Schema on read is obsolete. Welcome metaprogramming..pdf
Schema on read is obsolete. Welcome metaprogramming..pdfSchema on read is obsolete. Welcome metaprogramming..pdf
Schema on read is obsolete. Welcome metaprogramming..pdfLars Albertsson
 
Brighton SEO | April 2024 | Data Storytelling
Brighton SEO | April 2024 | Data StorytellingBrighton SEO | April 2024 | Data Storytelling
Brighton SEO | April 2024 | Data StorytellingNeil Barnes
 
Mature dropshipping via API with DroFx.pptx
Mature dropshipping via API with DroFx.pptxMature dropshipping via API with DroFx.pptx
Mature dropshipping via API with DroFx.pptxolyaivanovalion
 
Unveiling Insights: The Role of a Data Analyst
Unveiling Insights: The Role of a Data AnalystUnveiling Insights: The Role of a Data Analyst
Unveiling Insights: The Role of a Data AnalystSamantha Rae Coolbeth
 
Data-Analysis for Chicago Crime Data 2023
Data-Analysis for Chicago Crime Data  2023Data-Analysis for Chicago Crime Data  2023
Data-Analysis for Chicago Crime Data 2023ymrp368
 

Recently uploaded (20)

Industrialised data - the key to AI success.pdf
Industrialised data - the key to AI success.pdfIndustrialised data - the key to AI success.pdf
Industrialised data - the key to AI success.pdf
 
B2 Creative Industry Response Evaluation.docx
B2 Creative Industry Response Evaluation.docxB2 Creative Industry Response Evaluation.docx
B2 Creative Industry Response Evaluation.docx
 
BigBuy dropshipping via API with DroFx.pptx
BigBuy dropshipping via API with DroFx.pptxBigBuy dropshipping via API with DroFx.pptx
BigBuy dropshipping via API with DroFx.pptx
 
Carero dropshipping via API with DroFx.pptx
Carero dropshipping via API with DroFx.pptxCarero dropshipping via API with DroFx.pptx
Carero dropshipping via API with DroFx.pptx
 
VIP Call Girls Service Charbagh { Lucknow Call Girls Service 9548273370 } Boo...
VIP Call Girls Service Charbagh { Lucknow Call Girls Service 9548273370 } Boo...VIP Call Girls Service Charbagh { Lucknow Call Girls Service 9548273370 } Boo...
VIP Call Girls Service Charbagh { Lucknow Call Girls Service 9548273370 } Boo...
 
Generative AI on Enterprise Cloud with NiFi and Milvus
Generative AI on Enterprise Cloud with NiFi and MilvusGenerative AI on Enterprise Cloud with NiFi and Milvus
Generative AI on Enterprise Cloud with NiFi and Milvus
 
Low Rate Call Girls Bhilai Anika 8250192130 Independent Escort Service Bhilai
Low Rate Call Girls Bhilai Anika 8250192130 Independent Escort Service BhilaiLow Rate Call Girls Bhilai Anika 8250192130 Independent Escort Service Bhilai
Low Rate Call Girls Bhilai Anika 8250192130 Independent Escort Service Bhilai
 
04242024_CCC TUG_Joins and Relationships
04242024_CCC TUG_Joins and Relationships04242024_CCC TUG_Joins and Relationships
04242024_CCC TUG_Joins and Relationships
 
VIP Call Girls in Amravati Aarohi 8250192130 Independent Escort Service Amravati
VIP Call Girls in Amravati Aarohi 8250192130 Independent Escort Service AmravatiVIP Call Girls in Amravati Aarohi 8250192130 Independent Escort Service Amravati
VIP Call Girls in Amravati Aarohi 8250192130 Independent Escort Service Amravati
 
Midocean dropshipping via API with DroFx
Midocean dropshipping via API with DroFxMidocean dropshipping via API with DroFx
Midocean dropshipping via API with DroFx
 
RA-11058_IRR-COMPRESS Do 198 series of 1998
RA-11058_IRR-COMPRESS Do 198 series of 1998RA-11058_IRR-COMPRESS Do 198 series of 1998
RA-11058_IRR-COMPRESS Do 198 series of 1998
 
꧁❤ Aerocity Call Girls Service Aerocity Delhi ❤꧂ 9999965857 ☎️ Hard And Sexy ...
꧁❤ Aerocity Call Girls Service Aerocity Delhi ❤꧂ 9999965857 ☎️ Hard And Sexy ...꧁❤ Aerocity Call Girls Service Aerocity Delhi ❤꧂ 9999965857 ☎️ Hard And Sexy ...
꧁❤ Aerocity Call Girls Service Aerocity Delhi ❤꧂ 9999965857 ☎️ Hard And Sexy ...
 
Ravak dropshipping via API with DroFx.pptx
Ravak dropshipping via API with DroFx.pptxRavak dropshipping via API with DroFx.pptx
Ravak dropshipping via API with DroFx.pptx
 
Call me @ 9892124323 Cheap Rate Call Girls in Vashi with Real Photo 100% Secure
Call me @ 9892124323  Cheap Rate Call Girls in Vashi with Real Photo 100% SecureCall me @ 9892124323  Cheap Rate Call Girls in Vashi with Real Photo 100% Secure
Call me @ 9892124323 Cheap Rate Call Girls in Vashi with Real Photo 100% Secure
 
Schema on read is obsolete. Welcome metaprogramming..pdf
Schema on read is obsolete. Welcome metaprogramming..pdfSchema on read is obsolete. Welcome metaprogramming..pdf
Schema on read is obsolete. Welcome metaprogramming..pdf
 
Brighton SEO | April 2024 | Data Storytelling
Brighton SEO | April 2024 | Data StorytellingBrighton SEO | April 2024 | Data Storytelling
Brighton SEO | April 2024 | Data Storytelling
 
Mature dropshipping via API with DroFx.pptx
Mature dropshipping via API with DroFx.pptxMature dropshipping via API with DroFx.pptx
Mature dropshipping via API with DroFx.pptx
 
Unveiling Insights: The Role of a Data Analyst
Unveiling Insights: The Role of a Data AnalystUnveiling Insights: The Role of a Data Analyst
Unveiling Insights: The Role of a Data Analyst
 
Data-Analysis for Chicago Crime Data 2023
Data-Analysis for Chicago Crime Data  2023Data-Analysis for Chicago Crime Data  2023
Data-Analysis for Chicago Crime Data 2023
 
Sampling (random) method and Non random.ppt
Sampling (random) method and Non random.pptSampling (random) method and Non random.ppt
Sampling (random) method and Non random.ppt
 

PR-122: Can-Creative Adversarial Networks

  • 1. visionNoob (Jaewon Lee) PR-122 CAN: Creative Adversarial Networks, Generating "Art" by Learning About Styles and Deviating from Style NormsAhmed Elgammal, Bingchen Liu, Mohamed Elhoseiny, Marian Mazzone, (2017) 1 https://arxiv.org/abs/1706.07068
  • 2. 2 ∗This paper is an extended version of a paper published on the eighth International Conference on Computational Creativity (ICCC), held in Atlanta, GA, June 20th-June 22nd, 2017. The Art and Artificial Intelligence Laboratory at Rutgers: https://sites.google.com/site/digihumanlab/home
  • 3. 3 Figure 1: Example of images generated by CAN. The generated images vary from simple abstract ones to complex textures and compositions. (256 x 256)
  • 4. 4 Figure 1: Example of images generated by CAN. The generated images vary from simple abstract ones to complex textures and compositions. (256 x 256)
  • 5. 5 If we teach the machine about art and art styles and force it to generate novel images that do not follow established styles, what would it generate?
  • 6. 6 Would it generate something that is aesthetically appealing to humans? CAN : Top ranked by human subjects 1
  • 7. 7 Would that be considered “art”? CAN : Top ranked by human subjects 2
  • 8. t 8 Art Generating Agent The agent tries to generate that is novel, but not too novel. https://www.psychologytoday.com/intl/blog/in-the-brain-the-beholder/201405/its-acquired-taste-how-knowledge-drives-aesthetics?amp https://news.artnet.com/market/google-inceptionism-art-sells-big-439352 too novel novel deepdream not nobel “dazzling, druggy, and creepy”
  • 10. t 10 Art Generating Agent The agent tries to generate that is novel, but not too novel. https://www.psychologytoday.com/intl/blog/in-the-brain-the-beholder/201405/its-acquired-taste-how-knowledge-drives-aesthetics?amp https://news.artnet.com/market/google-inceptionism-art-sells-big-439352 “dazzling, druggy, and creepy” too novel novel deepdream not nobel
  • 11. 11 Dataset WikiArt Dataset (2015) 81,449 paintings from 1,119 artists (from 15th~21st century) with style label https://www.wikiart.org/
  • 12. 12 GAN is Emulative and not Creative Recap - Vanilla GAN min 𝐺 max 𝐷 𝑉 𝐷, 𝐺 = 𝔼 𝑥~𝑝 𝑑𝑎𝑡𝑎 [log 𝐷 𝑥 ] + 𝔼 𝑧~𝑝 𝑧 [log(1 − 𝐷 𝐺(𝑧 ))] https://blog.openai.com/generative-models/
  • 13. 13 min 𝐺 max 𝐷 𝑉 𝐷, 𝐺 = 𝔼 𝑥~𝑝 𝑑𝑎𝑡𝑎 [log 𝐷 𝑥 ] + 𝔼 𝑧~𝑝 𝑧 [log(1 − 𝐷 𝐺(𝑧 ))] min 𝐺 max 𝐷 𝑉 𝐷, 𝐺 = 𝔼 𝑥, 𝑐~𝑝 𝑑𝑎𝑡𝑎 [log 𝐷𝑟 𝑥 + log 𝐷𝑐 𝑐 = 𝑐|𝑥 ] + 𝔼 𝑧~𝑝 𝑧 [log(1 − 𝐷𝑟 𝐺(𝑧 )) − 𝑘=1 𝐾 1 𝐾 log(𝐷𝑐 𝑐 𝑘 𝐺 𝑧 + (1 − 1 𝑘 ) log(1 − 𝐷𝑐(𝑐 𝑘|𝐺 𝑧 ) ] recap - Vanilla GAN loss 𝑐 : style label from the the data distribution 𝑝 𝑑𝑎𝑡𝑎 K : # of style classes 𝐷𝑟 〮 : discriminate between real art and generated images 𝐷𝑐(〮) : discriminate between different style categories and estimate posterior (i.e., 𝐷𝑐 𝑐 𝑘 〮 = 𝑝(𝑐 𝑘|〮)) style classification style ambiguity maximizes the entropy of p(c|x) CAN : Loss function
  • 14. 14 CAN : Loss function Training Discriminator Training Generator max 𝐷 𝑉 𝐷, 𝐺 = 𝔼 𝑥, 𝑐~𝑝 𝑑𝑎𝑡𝑎 [log 𝐷𝑟 𝑥 + log 𝐷𝑐 𝑐 = 𝑐|𝑥 ] + # For Training data 𝔼 𝑧~𝑝 𝑧 [log(1 − 𝐷 𝐺(𝑧 ))] # For Generated data 𝔼 𝑧~𝑝 𝑧 [log(1 − 𝐷 𝐺(𝑧 )) − 𝑘=1 𝐾 1 𝐾 log(𝐷𝑐 𝑐 𝑘 𝐺 𝑧 + (1 − 1 𝑘 ) log(1 − 𝐷𝑐(𝑐 𝑘|𝐺 𝑧 ) ] min 𝐺 𝑉 𝐷, 𝐺 = style classification style ambiguity
  • 15. 15
  • 16. 16 Experiments CAN implementation : https://github.com/mlberkeley/Creative-Adversarial-Networks
  • 17. Experiments 17 Basically… Assessing the creativity of artifacts generated by the machine is an open and hard question. Dataset : WikiArt dataset Baseline 1 : original DCGAN (64 x 64) Baseline 2 : original DCGAN (256 x 256) Baseline 3 : CAN (w.o. style ambiguity) (256 x 256) Proposed : CAN (256 x 256)
  • 18. 18 We also do not see any recognizable figures. Many of the images seems abstract. Is that simply because it fails to emulate the art distribution or Is it because it tried to generate novel images? Is it at all creative? Figure 1: Example of images generated by CAN. The generated images vary from simple abstract ones to complex textures and compositions. (256 x 256)
  • 19. 19 Qualitative Result DCGAN 64 x 64 DCGAN 256 x 256 CAN 256 x 256 (w.o. style ambiguity) CAN 256 x 256
  • 20. 20 Quantitative Result Experiment 1 (Amazone Mturk) test the ability of the system to generate art that human users could not distinguish from top creative art that is being generated by artists today Q1: Do you think the work is created by an artist or generated by a computer? The user has to choose one of two answers: artist or computer. Q2: The user asked to rate how they like the image in a scale 1 (extremely dislike) to 5 (extremely like).
  • 21. 21 Figure 8: Art Basel Set: a collection of 25 paintings selected from Art Basel 2017 art fair. Shamir, Lior, Jenny Nissel, and Ellen Winner. "Distinguishing between abstract art by artists vs. children and animals: Comparison between human and machine perception." ACM Transactions on Applied Perception (TAP) 13.3 (2016): 17
  • 22. 22 Quantitative Result Experiment 2 (Amazone MTurk) Q1 How do you like this image: 1-extremely dislike ~ 5-extremely like. Q2 Rate the novelty of the image: 1-extremely not novel, ~ 5-extremely novel. Q3 Do you find the image surprising: 1-extremely not surprising ~ 5-extremely surprising. Q4 Rate the ambiguity of the image. I find this image: 1-extremely not ambiguous ~ 5-extremely ambiguous. Q5 Rate the complexity of the image. I find this image: 1-extremely simple ~ 5-extremely complex Q6 Do you think the image is created by an artist or generated by computer?
  • 23. 23 Quantitative Result Experiment 3 judge aspects related to whether the images generated by CAN can be considered art Q1: As I interact with this painting, I start to see the artist’s intentionality: it looks like it was composed very intentionally. Q2: As I interact with this painting, I start to see a structure emerging. Q3: Communication: As I interact with this painting, I feel that it is communicating with me. Q4: Inspiration: As I interact with this painting, I feel inspired and elevated.
  • 24. 24 Quantitative Result Experiment 4 CAN vs sc-CAN (sophisticated art-educated subjects) evaluate the effect of adding the style ambiguity loss to the CAN model Q1 Which image do you think is more novel? Q2 Which image do you think is more aesthetically appealing? Result: CAN images are more novel than sc-CAN : 59.47% CAN images are more aesthetically appealing than sc-CAN : 60%
  • 25. Q&A 25 Figure 5: Example of images generated by CAN. Top: Images ranked high in “likeness” according to human subjects. Bottom: Images ranked the lowest by human subjects.