Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
Chef*mo: 動画が調理器具を制御して調理を支援するシステムの提案 / Chefmo
Search
Tsubasa Yumura
December 23, 2017
Research
1
1.4k
Chef*mo: 動画が調理器具を制御して調理を支援するシステムの提案 / Chefmo
2017年12月23日(土)にグランフロント大阪で開催されたEC46の発表資料。論文はこちら →
https://goo.gl/6oJHh4
Tsubasa Yumura
December 23, 2017
Tweet
Share
More Decks by Tsubasa Yumura
See All by Tsubasa Yumura
Human-Computer Interaction (HCI) 入門 #北海道LT大会 / HCI2023
yumu19
0
160
Tinkercadの電子回路シミュレータはいいぞ #北海道まったりLT大会 / Tinkercad
yumu19
1
3k
メイカーズ文化とシチズンサイエンス #JOSS2021 / Makers Culture and Citizen Science
yumu19
0
590
バーチャルSNS clusterを用いたイベント開催 / Event Organization with Virtual SNS “cluster”
yumu19
0
120
石川での6年間 / 6 years in Ishikawa
yumu19
0
290
NT札幌2020 in cluster 開催報告 #XRMTG / NT Sapporo 2020 in cluster
yumu19
0
80
Space Apps Challenge Tokyo 2020 開会式
yumu19
0
92
社会人博士のススメ
yumu19
2
7.3k
SpaceApps COVID-19 Challenge開会式
yumu19
0
140
Other Decks in Research
See All in Research
熊本から日本の都市交通政策を立て直す~「車1割削減、渋滞半減、公共交通2倍」の実現へ~@公共交通マーケティング研究会リスタートセミナー
trafficbrain
0
180
メールからの名刺情報抽出におけるLLM活用 / Use of LLM in extracting business card information from e-mails
sansan_randd
2
220
2024/10/30 産総研AIセミナー発表資料
keisuke198619
1
370
Practical The One Person Framework
asonas
1
1.8k
LLM時代にLabは何をすべきか聞いて回った1年間
hargon24
1
530
尺度開発における質的研究アプローチ(自主企画シンポジウム7:認知行動療法における尺度開発のこれから)
litalicolab
0
360
TransformerによるBEV Perception
hf149
1
580
Introducing Research Units of Matsuo-Iwasawa Laboratory
matsuolab
0
1.2k
外積やロドリゲスの回転公式を利用した点群の回転
kentaitakura
1
710
言語処理学会30周年記念事業留学支援交流会@YANS2024:「学生のための短期留学」
a1da4
1
260
湯村研究室の紹介2024 / yumulab2024
yumulab
0
350
Embers of Autoregression: Understanding Large Language Models Through the Problem They are Trained to Solve
eumesy
PRO
7
1.2k
Featured
See All Featured
Intergalactic Javascript Robots from Outer Space
tanoku
270
27k
Producing Creativity
orderedlist
PRO
341
39k
Mobile First: as difficult as doing things right
swwweet
222
9k
Building Better People: How to give real-time feedback that sticks.
wjessup
365
19k
Building a Modern Day E-commerce SEO Strategy
aleyda
38
7k
Responsive Adventures: Dirty Tricks From The Dark Corners of Front-End
smashingmag
251
21k
Save Time (by Creating Custom Rails Generators)
garrettdimon
PRO
28
900
How to Think Like a Performance Engineer
csswizardry
21
1.2k
Building Flexible Design Systems
yeseniaperezcruz
327
38k
Design and Strategy: How to Deal with People Who Don’t "Get" Design
morganepeng
127
18k
Learning to Love Humans: Emotional Interface Design
aarron
273
40k
CoffeeScript is Beautiful & I Never Want to Write Plain JavaScript Again
sstephenson
159
15k
Transcript
Chef mo * ౬ଜ ཌྷ†, ݹؘ ༎ࢠ†, দຊ ՚า‡, ᖒ
ͻ͞Έ†, ܀ݪ Ұو‡ɹ ಈը͕ௐཧث۩Λ੍ޚͯ͠ௐཧΛࢧԉ͢ΔγεςϜͷఏҊ † Chefˎmo ϓϩδΣΫτ ‡ ాक़େֶ ใॲཧֶձ ୈ46ճΤϯλςΠϯϝϯτίϯϐϡʔςΟϯάˏάϥϯϑϩϯτେࡕ
Chef mo * Chefˎmo 2
Chef mo * ຊൃදͷ֓ཁ • ྉཧಈըʹௐཧεΫϦϓτΛຒΊࠐΈɺಈըͷ༰ʹ ߹Θͤͯௐཧث۩Λ੍ޚͯ͠ௐཧࢧԉΛߦ͏γεςϜ ΛఏҊ • ίϯηϓτ࣮Λߦͬͨ
• ಈըͱIoTػثͷ࿈ܞϑϨʔϜϫʔΫsrt.jsΛ༻͍ͨ • ʮ͔͚ΔʯʮࠞͥΔʯʮΔʯͷ3छྨͷௐཧߦಈΛ ࣮ͨ͠ • ٞͱ՝Λ·ͱΊͨ 3
Chef mo * ͡Ίʹ • ྉཧಈը • Ωϡʔϐʔ3ΫοΩϯάʢ1962ʙʣ • 1ྉཧಈըʢKurashiruͳͲʣ
4 [2]kurashiru http://www.kurashiru.com/ ޘָ ࢹௌΛָ͠Ή ެ։Λָ͠Ή ޘָ ࣮ӹ ྉཧʹؔ͢Δ ใΛಘΔ ࣮ӹ ྉཧಈըΛͳͥ؍Δ͔ʁ
Chef mo * ͡Ίʹ • ػցʹΑΔௐཧࢧԉٕज़ͱΈ߹ΘͤΔ͜ͱʹΑΓɺ ྉཧಈըͷࢹௌΛΑΓΑ͍ͷʹঢ՚Ͱ͖ΔͷͰͳ ͍͔ʁ 5 [4]Moley
[6]Cooky
Chef mo * ͡Ίʹ 6 ྉཧಈըͷ ݩͷׂ ௐཧࢧԉ ྟײ ָ͠͞
Chef mo * ಈը͕ௐཧث۩Λ੍ޚͯ͠ௐཧΛࢧԉ͢ΔγεςϜ ޘָ ࢹௌΛָ͠Ή ެ։Λָ͠Ή ޘָ ࣮ӹ ྉཧʹؔ͢Δ ใΛಘΔ ࣮ӹ Chefˎmo Ͱ֦ு
Chef mo * ؔ࿈ݚڀ | ػցʹΑΔௐཧࢧԉ 7 [4]Moley – The
world's first robotic kitchen http://www.moley.com/
Chef mo * ؔ࿈ݚڀ | ػցʹΑΔௐཧࢧԉ 8 [6]Sugiura et al.,
Cooking with robots: designing a household system working in open environments (2010)
Chef mo * ؔ࿈ݚڀ | ϨγϐͱWeb 9 [10]٢ͱٶԼ, άϥϑΟΧϧσʔλϑϩʔʹΑΔ ௐཧϨγϐϓϩάϥϛϯάݴޠͷఏҊ
(2010) [9] RecipeML • Ϩγϐ • ྉཧΛͭ͘ΔͨΊʹඞཁͳࡐྉௐཧํ๏ͳͲͷใΛॻ͖هͨ͠खॱॻ • खॱॻͱ͍͏ಛੑ͔ΒߏԽ͞Εͨهड़ʹ͍͍ͯΔ
Chef mo * ؔ࿈ݚڀ | ϨγϐͱWeb 10 [12]ᬑΒ, smoon: Webͷ࣮ମԽʹΑΔߦಈࢧԉͱͦͷࢼ࡞
(2011)
Chef mo * ؔ࿈ݚڀ | ಈըʹΑΔεΫϦϓτ࣮ߦ 11 [13]܀ݪͱڮຊ, srt.js: ө૾ίϯςϯπͷ
IoT ࢦ֦ு ϓϩάϥϜຒΊࠐΈϑϨʔϜϫʔΫ (2016) • srt.js • srtϑΝΠϧ(ࣈນ)ʹJavaScriptΛهड़࣮ͯ͠ߦ͢ΔϑϨʔϜϫʔΫ • ࠶ੜͨ͠ಈըͷҙͷλΠϛϯάʹ߹ΘͤͯίʔυΛ࣮ߦ
Chef mo * Chefˎmo • ௐཧεΫϦϓτ • ಈըʹ͋ΘͤͨௐཧࢧԉͷखॱΛهࡌ • User
Generated Contents(UGC)ͱͯ͠ͷੜΛఆ • ΩονϯϚωʔδϟ • ֤ՈఉͷΩονϯʹஔ • ௐཧث۩σΟεϓϨΠͳͲͷػثશମΛ౷ׅ 12
Chef mo * Chefˎmo • ௐཧࢧԉͷܗଶ 13
Chef mo * Chefˎmo • ௐཧࢧԉͷ߹͍ 14 ChefˎmoͷλʔήοτྖҬ
Chef mo * ίϯηϓτ࣮ • YouTubeͷطଘௐཧಈըΛ༻ • srt.jsͷϑϨʔϜϫʔΫΛར༻ͯ͠ௐཧεΫϦϓτΛ࣮ߦ • Webmo[16]Λར༻ͯ͠ௐཧث۩Λࣗ࡞
• ʮ͔͚ΔʯʮࠞͥΔʯʮΔʯΛ࣮ • 3ͭͷϢʔεέʔεΛ࣮ࢪ • (a)υϨογϯάΛ͔͚Δ • (b)ཛΛ༹͘ • (c)౾ΛΔ 15
Chef mo * ίϯηϓτ࣮ 16
Chef mo * ίϯηϓτ࣮ 17
Chef mo * ٞ | Chefˎmoͷ׆༻ • λϨϯτΩϟϥΫλʔͷጪґ • ༗໊γΣϑͷௐཧಈը
• ྉཧͷັྗΛߴΊΔߴڃ৯ࡐͷಈըͳͲͷಈը • өըΞχϝͱͷίϥϘϨʔγϣϯͷ࣮ݱ 18 [18]͚ͷϑϨϯζCAFE https://twitter.com/japaricafe_hara/status/929896900525830144
Chef mo * ٞ | ௐཧث۩ • ຊൃදͰWebmo(αʔϘϞʔλ)Λར༻ͯࣗ͠࡞ • API੍ޚͰ͖Δௐཧث۩ࠓޙීٴ͢ΔͰ͋Ζ͏
• ͍··Ͱʹͳ͍ௐཧث۩ͷࣗ࡞·Γͭͭ͋Δ • Maker Faire TokyoͰʮ৯ʯΧςΰϦ͕ઃஔ 19 [22];Γ͔͚ϓϩολ:LunchBot [24]ຊ֨ͩ͠ϝʔΧʔʮBUSHIʯ [25]ΫϨʔϓϩϘοτʮΫϨϓʯ
Chef mo * ٞ | ௐཧεΫϦϓτͷ࡞ • ຊൃදͷίϯηϓτ࣮Ͱsrt.jsʹ࣮ͯߦ͢ΔJavaScriptͰ WebmoͷؔΛهड़ •
جຊతͳௐཧߦಈΛϥΠϒϥϦͱͯ͠ཧ͠ػثͷࠩҟΛٵऩ • Ϩγϐͷղੳߦಈೝ͔ࣝΒௐཧεΫϦϓτͷࣗಈੜ 20 [28]Hashimoto et al., Kusk dataset: Toward a direct understanding of recipe text and human cooking activity (2014)
Chef mo * ٞ | ཧੈքͱಈըͷ༥߹ • ChefˎmoͷޮՌΛߴΊΔͨΊʹཧੈքͱಈըͷ༥߹͕ॏཁ • VR/MR/ARͷ׆༻
• ಈըͱͷଓײ 21 [29]ᬑͱதଜ, WorldConnector: Χϝϥͷମੑ༩ʹΑΔ ө૾ੈքೖΓࠐΉΠϯλϑΣʔε (2014)
Chef mo * ՝ • ௐཧߦಈͷཻ • ଟ͘ͷௐཧߦಈʹલ४උ͕ඞཁ • ྫ͑ʮΔʯͷલʹʮ৯ࡐΛ·ͳ൘ͷ্ʹͷͤΔʯͱ͍
͏४උߦಈ͕͋Δ • ղܾҊ • ௐཧεΫϦϓτৄࡉԽʹΑΔࣗಈԽ • ਓ͕ڠௐͰ͖ΔΑ͏ࢧ࣋ɾ༠ಋ 22
Chef mo * ՝ • ಈըͷ࣌ؒ࣠ͷௐ • طଘͷௐཧಈըਓ͕ࢹௌ͢ΔͨΊʹͭ͘ΒΕ͍ͯ Δ •
ಈըͰෆཁͳ෦লུ͞ΕΔͨΊ࣮ࡍͷௐཧ࣌ ؒΑΓ͍ • ղܾҊ • ௐཧߦಈͷਐḿʹΑΓಈը࠶ੜΛ੍ޚ 23
Chef mo * ·ͱΊ • ಈը͕ௐཧΛࢧԉ͢ΔγεςϜChefˎmoͷίϯηϓτ Λఏএ • Chefˎmoͷ׆༻ํ๏՝ɺՄೳੑʹ͍ͭͯٞΛ· ͱΊͨ
• ௐཧث۩ͷൃలಈըࢹௌελΠϧͷมԽʹΑͬͯ Chefˎmoͷ׆༻໘͕Δ 24