Fashion Tech x Machine Learning/twm_fashion_ml

by tn1031

Slide 1

Slide 1 text

Fashion Techͷ͋Μͪΐ͜ -ػցֶशฤ- 2016/4/16 TokyoWebMining #53 @tn1031

Slide 2

Slide 2 text

Agenda 1. Fashion x Technologyͷࣄྫ঺հ 2. Fashion෼໺ͷݚڀࣄྫ 3. ·ͱΊ 2

Slide 3

Slide 3 text

ࣗݾ঺հ ▸ தଜ ୓ຏ / @tn1031 ▸ σʔλαΠΤϯςΟετ ▸ SIer(2೥൒) -> VASILY ▸ ػցֶशΛઐ߈ ▸ ϊϯύϥϝτϦοΫϕΠζ ▸ ม෼ϕΠζ ▸ ϊϯύϥຊָ͕͠Έ ▸ TokyoWebMiningͷొஃ͸2ճ໨ ▸ લճ͸ɺը૾ೝࣝʹ͍ͭͯൃද 3 @tn1031 ਓ޻஌ೳ͸ɹɹɹɹɹ झຯͰᅂΉఔ౓

Slide 4

Slide 4 text

VASILYͱiQONʹ͍ͭͯ

Slide 5

Slide 5 text

VASILYͱiQONʹ͍ͭͯ ϑΝογϣϯΞϓϦʮiQONʯΛӡӦ͍ͯ͠·͢ 5

Slide 6

Slide 6 text

VASILYͱiQONʹ͍ͭͯ Ϣʔβ͕ϑΝογϣϯΞΠςϜ Λ૊Έ߹ΘͤͯίʔσΛ࡞ΕΔ 6

Slide 7

Slide 7 text

VASILYͱiQONʹ͍ͭͯ ؾʹೖͬͨΞΠςϜ͸ͦͷ৔ͰECαΠτʹඈΜͰ͓ങ͍෺Ͱ͖Δ 7 ఏܞECαΠτ

Slide 8

Slide 8 text

VASILYͱiQONʹ͍ͭͯ ೔ຊதͷECαΠτͷ঎඼৘ใΛΫϩʔϦϯά 8

Slide 9

Slide 9 text

VASILYͱiQONʹ͍ͭͯ iQON͕ѻ͏σʔλͷྫɿίʔσΟωʔτ 9 ίʔσ ‣ ࢖ΘΕͨΞΠςϜ ‣ ΞΠςϜͷϒϥϯυ ‣ ϨΠΞ΢τ ‣ λΠτϧ ‣ ίϝϯτ ‣ λά ‣ ɾɾɾ

Slide 10

Slide 10 text

Fashion x Technologyͷࣄྫ঺հ

Slide 11

Slide 11 text

Fashion x Technologyʁ ‣ ϑΝογϣϯۀք͸ɺ ‣ Ϗδωεͷن໛͕େ͖͍ ‣ ੜ׆ʹ͔ܽͤͳ͍࢈ۀ ‣ IT͕ਁಁ͍ͯ͠ͳ͍ 11 ϑΝογϣϯۀքʹɹɹɹɹɹ ITͷ೾͕དྷͯΔΒ͍͠

Slide 12

Slide 12 text

Fashioning Data: A 2015 Update Data Innovations from the Fashion Industry 12 http://www.oreilly.com/data/free/fashioning-data.csp Έ͍ͨͳ͜ͱ͕ॻ͍ͯ͋Δຊ

Slide 13

Slide 13 text

Fashion x Technologyͷࣄྫ ςΫϊϩδʔΛ׆͔ͨ͠αʔϏε͕ଟ਺ଘࡏ͢Δ 13 WHERETOGET ‣ http://wheretoget.it/ Styloko ‣ https://www.styloko.com/ deepomatic ‣ https://www.deepomatic.com/

Slide 14

Slide 14 text

ࣄྫ1:WHERETOGE ‣ ϑΝογϣϯΞΠςϜͷը૾Λೖྗ͢ΔͱɺͦͷΞΠςϜ͕Ͳ͜Ͱख ʹೖΔͷ͔ڭ͑ͯ͘ΕΔαʔϏε ‣ ػց͕ݕࡧ͢ΔͷͰ͸ͳ͘ɺ஌͍ͬͯΔਓ͕ڭ͑ͯ͘ΕΔ 14

Slide 15

Slide 15 text

ࣄྫ2:Styloko ‣ ը૾Λೖྗ͢Δͱɺ৭΍ฑ͕Α͘ࣅͨΞΠςϜΛ୳ͯ͘͠ΕΔ ‣ ೖྗ͸(ͨͿΜ)෰Ͱͳͯ͘΋ྑ͍ 15

Slide 16

Slide 16 text

ࣄྫ3:deepomatic ‣ ը૾͔ΒϑΝογϣϯ ΞΠςϜΛݕग़͠ɺը ૾ݕࡧͰಉ͡ΞΠςϜ Λฦ͢ 16

Slide 17

Slide 17 text

Fashion x Technologyͷࣄྫ ‣ ը૾Λѻ͏αʔϏε͕ଟ͍ ‣ ෰Λങ͏্Ͱ͔ܽͤͳ͍ཁૉ ‣ ݕࡧ΍ύʔιφϥΠζͷधཁ͕͋Δ ‣ େྔͷΞΠςϜ͔Β໨తͷΞΠςϜΛ୳͢ ‣ ݸਓͷझ޲Λ൓өͨ͠ϥϯΩϯά 17 ϑΝογϣϯͱػցֶशͷ ਌࿨ੑ͸େ͖͍

Slide 18

Slide 18 text

Fashion෼໺ͷݚڀࣄྫ

Slide 19

Slide 19 text

FashionΛର৅ʹͨ͠ݚڀͰѻΘΕΔ໰୊ 19 ը૾ܥͷख๏͸ϑΝογϣϯσʔλͱͷ਌࿨ੑ͕ߴ͍ɻҎԼͷ3߲໨͕ओྲྀͳҹ৅ɻ ΞΠςϜݕग़ ύʔε ྖҬ෼ׂ ϑΝογϣϯ ΞΠςϜݕࡧ Ϩίϝϯυ ςʔϚ λεΫ ϑΝογϣϯྖҬ΁ͷద༻ ը૾͔ΒΞΠςϜʹ֘౰͢ΔՕॴΛಛఆ͠ɺ ύʔε΍ΧςΰϦͷ൑ผΛߦ͏ ѻ͏ը૾͕ݶఆ͞ΕΔҝɺ ࣄલ஌ࣝΛ׆༻Ͱ͖Δ ΫΤϦʹҰகɾྨࣅ͢Δ ΞΠςϜΛఏࣔ͢Δ ΤοδΛ༻͍ͨΞΠςϜͷ۠ผ͕ࠔ೉ͳҝɺ ৭΍ฑ΍ૉࡐͷ৘ใΛ൓өͰ͖ΔϞσϧ͕ ඞཁ ߦಈσʔλ΍ΞΠςϜؒྨࣅ౓Λར༻ͯ͠ɺ Ϣʔβ΁ͷΞΠςϜఏࣔΛ࠷దԽ͢Δ ߦಈσʔλͷଞʹΞΠςϜͷը૾ಛ௃ྔ΍ ྲྀߦΛϞσϧʹ૊ΈࠐΉ͜ͱ͕Ͱ͖Δ

Slide 20

Slide 20 text

ΞΠςϜݕग़ ‣ http://arxiv.org/abs/1411.5319 ‣ http://arxiv.org/abs/1603.07063 ‣ http://vision.is.tohoku.ac.jp/~kyamagu/papers/yamaguchi_cvpr2012.pdf 20 ը૾͔ΒΞΠςϜʹ֘౰͢ΔྖҬɾۣܗΛ੾Γग़͠ɺΧςΰϦ൑ผΛߦ͏ɻ ΞΠςϜ͕ݱΕΔ࠲ඪʹ͸نଇੑ͕͋Γɺ͜ͷੑ࣭ΛͲͷ༷ʹֶशʹར༻͢Δ͔͕ϙΠϯτɻ

Slide 21

Slide 21 text

Fashion Apparel Detection: The Role of Deep Convolutional Neural Network and Pose-dependent Priors (WACV 2016) ‣ ϑΝογϣϯΞΠςϜͷݕग़ ‣ ண͍ͯΔ΋ͷ͚ͩͰͳ͘ɺ਎ʹ෇͚Δ΋ͷ΋ର৅ɹɹɹɹɹɹɹɹɹ (hat, glasses, bag, pants, shoes and so on.) ‣ ֶशʹDeep Convolutional Neural NetworkΛར༻ ‣ ֤ΞΠςϜ͕ग़ݱ͠΍͍͢ҐஔΛࣄલ஌ࣝͱֶͯ͠शʹ׆༻͢Δ࢓૊Έ ΛఏҊ 21 ֓ཁ Fashion Apparel Detection: The Role of Deep Convolutional Neural Network and Pose-dependent Priors http://arxiv.org/abs/1411.5319

Slide 22

Slide 22 text

Fashion Apparel Detection: The Role of Deep Convolutional Neural Network and Pose-dependent Priors (WACV 2016) 22 Contribution ‣ domain-speciﬁc priorsͳࣄલ৘ใΛ׆༻ֶͨ͠शํ๏ ΛఏҊͨ͠ ‣ ϋϯυόοά͸खट΍खͷۙ͘ʹ͋Γɺۺ͸଍ͷۙ͘ʹग़ݱ͠ қ͍ ‣ ΞΠςϜͷେ͖͞͸͍ࣸͬͯΔਓͷେ͖͞ʹൺྫ͢Δ ‣ ϙʔζݕग़ΛߦͬͯΞΠςϜͷݕग़ਫ਼౓ΛߴΊͨ ‣ ΞΠςϜީิྖҬͷ࠲ඪͱϙʔζݕग़ͷग़ྗΛjoint͍ͤͯ͞Δ

Slide 23

Slide 23 text

Fashion Apparel Detection: The Role of Deep Convolutional Neural Network and Pose-dependent Priors (WACV 2016) 23 Overview of the proposed algorithm ‣ ΞΠςϜީิྖҬ͔Βͷಛ௃நग़͸CNNͰߦ͏ ‣ ͦΕͧΕͷΫϥεͷग़ݱ֬཰Λ1-vs-rest SVMsͰݟੵ΋Δ ‣ ϙʔζਪఆʹجͮ͘Ґஔ৘ใʹΑͬͯ֬཰Λิਖ਼͢Δ ‣ Non-Maximum suppressionͰ੔͑ͯ࠷ऴతͳग़ྗͱ͢Δ

Slide 24

Slide 24 text

Fashion Apparel Detection: The Role of Deep Convolutional Neural Network and Pose-dependent Priors (WACV 2016) 24 Model Object Proposal ‣ ը૾͔ΒΞΠςϜྖҬͷީิΛऔಘ͢Δ ‣ ͦΕͧΕͷީิྖҬʹ͍ͭͯΞΠςϜͷ༗ແΛ൑ผ ثͰॲཧͯ͠false positiveΛݮΒ͢ 0CKFDU1SPQPTBM *NBHF'FBUVSF&YUSBDUJPO 47.USBJOJOH 1SPCBCJMJTUJDGPSNVMBUJPO "QQFBSBODFCBTFE1PTUFSJPS (FPNFUSJD1SJPST Image Feature Extraction ‣ ֶशࡁΈCaffeNetΛ༻͍Δ ‣ ग़ྗ૚ͷखલͷग़ྗ4096࣍ݩͷϕΫτϧΛಛ௃ྔͱ ͢Δ

Slide 25

Slide 25 text

Fashion Apparel Detection: The Role of Deep Convolutional Neural Network and Pose-dependent Priors (WACV 2016) 25 Model Appearance-based Posterior ‣ SVMͷύϥϝʔλΛར༻ͯ͠posteriorΛܭࢉ͢Δ ‣ ɹ ‣ λ͕൑ผਫ਼౓ʹޮ͍ͯ͘ΔͷͰɺ࠷ऴతͳ݁Ռʹجͮ ͖λΛνϡʔχϯά͢Δ 0CKFDU1SPQPTBM *NBHF'FBUVSF&YUSBDUJPO 47.USBJOJOH 1SPCBCJMJTUJDGPSNVMBUJPO "QQFBSBODFCBTFE1PTUFSJPS (FPNFUSJD1SJPST SVM training ‣ ΞΠςϜͷΫϥε(bag,hat,...)ͦΕͧΕʹ͍ͭͯઢܗ SVMͰ2஋൑ผΛߦ͏

Slide 26

Slide 26 text

Fashion Apparel Detection: The Role of Deep Convolutional Neural Network and Pose-dependent Priors (WACV 2016) 26 Model 0CKFDU1SPQPTBM *NBHF'FBUVSF&YUSBDUJPO 47.USBJOJOH 1SPCBCJMJTUJDGPSNVMBUJPO "QQFBSBODFCBTFE1PTUFSJPS (FPNFUSJD1SJPST Geometric Priors ‣ Ґஔ৘ใʹؔ͢ΔpriorΛ2छྨఆٛ͢Δ ‣ ۣܗͷप௕ʹ͍ͭͯɺ ‣ ۣܗͷΞεϖΫτൺʹ͍ͭͯɺ ‣ ۣܗͷத৺࠲ඪʹ͍ͭͯɺ Probabilistic formulation ‣ Appearance-based PosteriorͱGeometric Priors͔ ΒPosteriorΛܭࢉ͢Δ

Slide 27

Slide 27 text

Fashion Apparel Detection: The Role of Deep Convolutional Neural Network and Pose-dependent Priors (WACV 2016) 27 Result

Slide 28

Slide 28 text

ݕࡧ ‣ http://cseweb.ucsd.edu/~chunbinlin/papers/www16demo.pdf ‣ http://arxiv.org/pdf/1505.07922.pdf ‣ http://www.iis.sinica.edu.tw/papers/song/18378-F.pdf 28 طଘͷΞϧΰϦζϜΛϑΝογϣϯʹಛԽͤ͞ΔͨΊɺฑ΍ૉࡐͳͲͷ෇Ճ৘ใΛ༻͍Δɻ ΞΠςϜؒͷؔ܎Λฏ໘ʹϚοϐϯάͯ͠ΈΔͱָ͍͠ɻ

Slide 29

Slide 29 text

Fashionista: A Fashion-aware Graphical System for Exploring Visually Similar Items (WWW 2016) ‣ ϑΝογϣϯΞΠςϜͷݕࡧγεςϜ ‣ ࢹ֮తʹྨࣅ͍ͯ͠ΔΞΠςϜͷݕࡧγεςϜΛ։ൃͨ͠ ‣ graphical interfaceʹͩ͜ΘΓ͕͋Δ ‣ ྲྀߦΓΛ౿·্͑ͨͰࣅ͍ͯΔΞΠςϜΛఏࣔ͢Δ ‣ ֶशʹ༻͍Δ৘ใ ‣ ը૾ಛ௃ྔ ‣ AmazonͷߪങσʔλͱλΠϜελϯϓ 29 ֓ཁ Fashionista: A Fashion-aware Graphical System for Exploring Visually Similar Items http://cseweb.ucsd.edu/~chunbinlin/papers/www16demo.pdf

Slide 30

Slide 30 text

Fashionista: A Fashion-aware Graphical System for Exploring Visually Similar Items(WWW 2016) 30 Contribution ‣ ࢹ֮తʹྨࣅ͍ͯ͠ΔΞΠςϜΛ୳ͤΔgraphical interface ”Fashionista”Λ։ൃͨ͠ ‣ ࢹ֮తͳ৘ใ(ը૾ಛ௃ྔ)Λྨࣅ౓ܭࢉʹར༻͢Δ ‣ ”Fashionista”͸τϨϯυΛՄࢹԽͯ͠ࠓྲྀߦ͍ͬͯΔΞΠςϜΛ୳͠ ΍͍ͯ͘͢͠Δ ‣ Amazon͔Β11೥ʹΘͨͬͯɺ0.6 million items, 1.8 million users, and 3.2 million user-item

Slide 31

Slide 31 text

Fashionista: A Fashion-aware Graphical System for Exploring Visually Similar Items(WWW 2016) 31 A screenshot of Fashionista http://132.239.95.211:8080/demowww/index.jsp

Slide 32

Slide 32 text

Fashionista: A Fashion-aware Graphical System for Exploring Visually Similar Items(WWW 2016) 32 Architecture of Fashionista

Slide 33

Slide 33 text

Fashionista: A Fashion-aware Graphical System for Exploring Visually Similar Items(WWW 2016) 33 Fashion Learner 5SBJOJOH%BUB ‣ Amazonͷߪങσʔλɹɹ (user, item, timestamp) ‣ pretrained CNN͔Βɹɹɹɹ ը૾ಛ௃ྔΛநग़ 'BTIJPO-FBSOFS ,OPXMFEHF ‣ ΞΠςϜಉ͕࢜ࣅ͍ͯΔʗࣅͯ ͍ͳ͍ͷ৘ใ ‣ tŖSNEͰѹॖͯ͠ฏ໘ʹɹ Ϛοϐϯά ‣ ΞΠςϜͷ͓͠ΌΕ౓ͷਪҠ ‣ Matrix Factorizationϕʔεͷख ๏ͰɺߦಈσʔλʹՃ͑ͯը૾ ಛ௃ྔΛར༻͢Δ ‣ ҎԼͷ࣌ܥྻ৘ใΛѻ͏ ‣ ΞΠςϜͷັྗʹ͍ͭͯͷ࣌ ؒਪҠ ‣ ࢹ֮త৘ใʹର͢ΔϢʔβ ͷ൓ԠͷਪҠ ‣ Ϣʔβຖͷᅂ޷ͷมԽ

Slide 34

Slide 34 text

Ϩίϝϯυ ‣ http://www.lv-nus.org/papers/2012/magic_closet-MM12.pdf ‣ http://labs.ebay.com/sites/default/ﬁles/14_fashionrecommender_kdd_ebayresearchlabs.pdf ‣ http://arxiv.org/abs/1506.04757 34 ߦಈσʔλ͚ͩͰͳ͘ɺΞΠςϜͷը૾ಛ௃ྔ΍ྲྀߦΛ൓өͤͨ͞ΞϧΰϦζϜ͕ఏҊ͞Ε͍ͯΔɻ Ϣʔβͷझ޲͚ͩͰͳ͘ɺΞΠςϜಉ࢜ͷ૬ੑΛߟྀͨ͠γεςϜ΋ଘࡏ͢Δɻ

Slide 35

Slide 35 text

“Hi, Magic Closet, Tell Me What toWear!”(ACM MM 2012) ‣ occasionʹର͢ΔϨίϝϯσʔγϣϯ ‣ Wedding΍σʔτͱ͍ͬͨoccasionΛೖྗ͢Δͱɺࣗ਎ ͷϑΥτΞϧόϜ΍ΦϯϥΠϯγϣοϓ͔Βద੾ͳ෰૷Λ ୳ͯ͠ఏҊͯ͘͠ΕΔ ‣ Magic Closet͕ղ͘໰୊͸2छྨ ‣ wear properly(ଐੑ)ɹɹɹɹɹɹɹɹɹɹɹɹɹɹɹɹ ʮόϯέοτʹ͸ύϯπεʔπΑΓυϨε͕ɹద͍ͯ͠Δʯ ‣ wear aesthetically(ඒ͠͞)ɹɹɹɹɹɹɹɹɹɹɹɹɹɹ ʮ੺͍Tγϟπʹ͸྘ΑΓനͷύϯπ͕ద͍ͯ͠Δʯ 35 ֓ཁ “Hi, Magic Closet, Tell Me What to Wear!" http://www.lv-nus.org/papers/2012/magic_closet-MM12.pdf

Slide 36

Slide 36 text

“Hi, Magic Closet, Tell Me What toWear!”(ACM MM 2012) 36 Magic Closetͷ࢓૊Έ

Slide 37

Slide 37 text

“Hi, Magic Closet, Tell Me What toWear!”(ACM MM 2012) 37 Magic Closetͷ࢓૊Έ TDFOBSJPɿPDDBTJPOΛࢦఆ ʮ*BNGPSBDPOGFSFODF QMFBTF SFDPNNFOETPNFUIJOHGSPNNZ QIPUPBMCVN ʯ TDFOBSJPɿPDDBTJPOͱ্൒਎ͷ෰Λࢦఆ ʮ*BNHPJOHPOBEBUF QMFBTFpOE TPNFUIJOHUPNBUDIXJUIUIJT5 TIJSU ʯ

Slide 38

Slide 38 text

“Hi, Magic Closet, Tell Me What toWear!”(ACM MM 2012) 38 Magic Closetͷ࢓૊Έ ز͔ͭͷ؍఺͔ΒϨίϝϯυΛߦ͏ B "UUSJCVUFWT0DDBTJPO.PEFM ‣ ෰ͷଐੑͱPDDBTJPOͷ૬ੑ C "UUSJCVUFWT"UUSJCVUF.PEFM ‣ ෰ͷଐੑಉ࢜ͷ૬ੑ ্൒਎ͱԼ൒ ਎ D $MPUIJOH"UUSJCVUF&TUJNBUJPO ‣ ෰ͷଐੑͷਪఆ

Slide 39

Slide 39 text

“Hi, Magic Closet, Tell Me What toWear!”(ACM MM 2012) 39 Magic Closetͷ࢓૊Έ ݁Ռͷग़ྗ PDDBTJPOΛࢦఆͨ͠৔߹ ྫɿDPOGFSFODF ‣ ద੾ͳ෰ ͷ૊Έ߹Θͤ ΛఏҊ PDDBTJPOͱ্൒਎ͷ෰Λࢦఆͨ͠৔߹ ྫɿ EBUF 5TIJSU ‣ ࢦఆͨ͠෰͓ΑͼPDDBTJPOͱ૬ੑͷྑ͍ ෰ΛఏҊ

Slide 40

Slide 40 text

“Hi, Magic Closet, Tell Me What toWear!”(ACM MM 2012) ‣ ҎԼͷ2ͭͷ໰୊ʹ࠷ॳʹऔΓ૊Μͩ ‣ ϑΥτΞϧόϜ͔Βoccasionʹదͨ͠෰૷Λ୳ͯ͠ਪન͢Δ ‣ ΫΤϦͷΞΠςϜ΍occasionʹ૬ੑͷྑ͍ΞΠςϜΛECαΠτ͔Β୳͢ ‣ What-to-Wear (WoW)σʔληοτΛߏஙͨ͠ ‣ 24,417ຕɺ౰࣌࠷େڃΒ͍͠ ‣ occasionͱ෰ͷ૬ੑΛಉ࣌ʹֶश͢Δlatent SVMΛఏҊͨ͠ 40 Contribution

Slide 41

Slide 41 text

“Hi, Magic Closet, Tell Me What toWear!”(ACM MM 2012) ‣ ֶशσʔλ ‣ ‣ :্൒਎/Լ൒਎ͷը૾ಛ௃ྔ ʗ :্൒਎ͷଐੑ(৭ͱ͔)ʗ :Լ൒਎ͷଐੑ ʗ :occasionΧςΰϦ 41 Model ( x, au, al, o ) ‣ Recommendation function x al au o

Slide 42

Slide 42 text

“Hi, Magic Closet, Tell Me What toWear!”(ACM MM 2012) 42 Model 'FBUVSFWT0DDBTJPO ‣ ը૾ಛ௃ྔͱoccasion ‣ ઢܗϞσϧ 'FBUVSFWT"UUSJCVUF ‣ ը૾ಛ௃ྔͱ෰ͷଐੑ ‣ ઢܗϞσϧ "UUSJCVUFWT"UUSJCVUF ‣ ෰ͷଐੑͷ૊Έ߹Θͤ ͷ૬ੑ ‣ ্൒਎ͱԼ൒਎ͷϖΞ ʹ͍ͭͯܭࢉ͢Δ ‣ ্Լͷ෰ͷ૬ੑʹ͍ͭ ֶͯश͢Δ "UUSJCVUFWT0DDBTJPO ‣ ෰ͷଐੑͱoccasion ‣ ઢܗϞσϧ ‣ occasionʹదͨ͠෰૷ Λֶश͢Δ

Slide 43

Slide 43 text

“Hi, Magic Closet, Tell Me What toWear!”(ACM MM 2012) ‣ ӡ༻࣌ ‣ occasionʹదͨ͠෰ͷϨίϝϯυ ‣ Լ൒਎givenͰ্൒਎ͷ෰ͷϨίϝϯυ ‣ ͜͜Ͱɺ 43 Model x ⇤ = arg maxx 2 X t fw(x, o) x ⇤ u = arg maxxu 2 X t u fw([xu; xl], o) fw(x, o) = max au,al w T (x, au, al, o)

Slide 44

Slide 44 text

“Hi, Magic Closet, Tell Me What toWear!”(ACM MM 2012) 44 Result ΞΠςϜͷଐੑ ಉ࢜ͷ૬ੑ ଐੑͱoccasionͷ૬ੑ

Slide 45

Slide 45 text

“Hi, Magic Closet, Tell Me What toWear!”(ACM MM 2012) 45 Result

Slide 46

Slide 46 text

·ͱΊ

Slide 47

Slide 47 text

·ͱΊ ‣ Fashion x Technology͕੝Γ্͕͍ͬͯΔ ‣ TechnologyΛ׆༻ͨ͠αʔϏε͕ଟ਺ଘࡏ͢Δ ‣ ݚڀ։ൃ΋੝ΜʹߦΘΕ͍ͯΔ 47

Slide 48

Slide 48 text

͝ਗ਼ௌ͋Γ͕ͱ͏͍͟͝·ͨ͠

Slide 49

Slide 49 text

ࢀߟ ‣ http://cs229.stanford.edu/proj2009/McDanielsWorsley.pdf ‣ http://www.cv-foundation.org/openaccess/content_cvpr_2015/papers/Simo- Serra_Neuroaesthetics_in_Fashion_2015_CVPR_paper.pdf ‣ http://labs.ebay.com/sites/default/files/13_stylefinder_cvpr_ebayresearchlabs.pdf ‣ http://vision.is.tohoku.ac.jp/~kyamagu/papers/yamaguchi_cvpr2012.pdf ‣ http://arxiv.org/abs/1411.5319 ‣ http://labs.ebay.com/sites/default/files/14_fashionrecommender_kdd_ebayresearchlabs.pdf ‣ http://www.cv-foundation.org/openaccess/content_cvpr_2015/papers/ Chen_Deep_Domain_Adaptation_2015_CVPR_paper.pdf ‣ http://arxiv.org/abs/1603.07063 ‣ http://arxiv.org/abs/1506.04757 ‣ http://cseweb.ucsd.edu/~chunbinlin/papers/www16demo.pdf ‣ http://arxiv.org/pdf/1505.07922.pdf ‣ http://www.iis.sinica.edu.tw/papers/song/18378-F.pdf ‣ http://arxiv.org/abs/1602.01585 ‣ http://www.lv-nus.org/papers/2012/magic_closet-MM12.pdf ‣ http://arxiv.org/abs/1405.4013 49