Upgrade to Pro — share decks privately, control downloads, hide ads and more …
Sign up for free

Learning 
Cross-modal Embeddings
 for Cooking R...

Sponsored · Your Podcast. Everywhere. Effortlessly. Share. Educate. Inspire. Entertain. You do you. We'll handle the rest. →
Avatar for Leszek Rybicki Leszek Rybicki
August 06, 2017
Research
830
1

Learning 
Cross-modal Embeddings
 for Cooking Recipes 
and Food Images

Review and presentation of the Image to Recipe paper and model

Avatar for Leszek Rybicki

Leszek Rybicki

August 06, 2017

More Decks by Leszek Rybicki

See All by Leszek Rybicki
Let's talk about Fakes
Avatar for Leszek Rybicki lunardog
0
180
How to Patch Image Classifiers
Avatar for Leszek Rybicki lunardog
0
3k
Towards Realistic Predictors - EN
Avatar for Leszek Rybicki lunardog
0
2.9k
Towards Realistic Predictors
Avatar for Leszek Rybicki lunardog
1
2.7k
Deep Learning Hot Dog Detector
Avatar for Leszek Rybicki lunardog
0
330
Finding beans in burgers: paper reading notes
Avatar for Leszek Rybicki lunardog
0
2k
Kelner: Serve Your Models
Avatar for Leszek Rybicki lunardog
0
170
Image Analysis at Cookpad
Avatar for Leszek Rybicki lunardog
1
2.2k
Kelner: serve your models
Avatar for Leszek Rybicki lunardog
1
440

Other Decks in Research

See All in Research
SAKURAONE: An Open Ethernet-based AI HPC System And Its Observed Workload Dynamics in a Single-Tenant LLM Development Environment
Avatar for Yuuki Tsubouchi (yuuk1) yuukit
1
610
Karkada さんの論文 × 2 の紹介: (1) Closed-Form Training Dynamics Reveal Learned Features and Linear Structure in Word2Vec-like Models, (2) Symmetry in language statistics shapes the geometry of model representations
Avatar for Sho Yokoi eumesy
PRO
1
760
GLIM とMegaParticles：正規分布近似の限界とタイトカップリング＆パーティクルフィルタの進展 / GLIM and MegaParticles : Progress of the distribution representation in SLAM
Avatar for koide3 koide3
0
870
秋葉原ウォーカブル基礎調査報告書
Avatar for 日本大学 都市計画研究室（泉山ゼミ） izumiyama_lab
1
120
プレイス・ワークショップ秋葉原
Avatar for 日本大学 都市計画研究室（泉山ゼミ） izumiyama_lab
1
100
LLM の Attention 機構まとめ — 数式・計算量・メモリ
Avatar for プワー puwaer
8
2.7k
【ローカルAIに向き合う展示会vol.2】液体時間定数型モジュールを用いた オリジナルの双方向エンコーダーモデルNexteraBERT 推論速度向上検討並びにダウンストリーム評価
Avatar for Rikka Botan rikkabotan7
0
200
長時間動画QAにおけるマルチエージェント推論 ・SVAgent: Storyline-Guided Long Video Understanding via Cross-Modal Multi-Agent Collaboration
Avatar for 村川卓也 murakawatakuya
1
210
最先端NLP勉強会2026 論文紹介：Reasoning with Sampling: Your Base Model is Smarter Than You Think (ICLR 2026 paper)
Avatar for Kogoro kogoro
4
630
Language and AI
Avatar for Ayana Niwa ayaniwa
0
290
PGDM: Physically Guided Diffusion Model for L Downscaling
Avatar for SatAI.challenge satai
3
490
【Zozo Research 技術共有会】三次元領域の現在と展望
Avatar for 小島瑞貴 mickey_0226
3
610

Featured

See All Featured
How to Talk to Developers About Accessibility
Avatar for Jason CranfordTeague jct
2
550
ラッコキーワード サービス紹介資料
Avatar for ラッコ株式会社 rakko
1
5M
Introduction to Domain-Driven Design and Collaborative software design
Avatar for Kenny Baas-Schwegler baasie
1
990
For a Future-Friendly Web
Avatar for Brad Frost brad_frost
183
10k
Navigating Team Friction
Avatar for Lara Hogan lara
192
16k
Optimizing for Happiness
Avatar for Tom Preston-Werner mojombo
378
71k
So, you think you're a good person
Avatar for Per Axbom axbom
PRO
2
2.2k
Reflections from 52 weeks, 52 projects
Avatar for Jefferson Lam jeffersonlam
356
21k
Side Projects
Avatar for Sacha Greif sachag
456
43k
Visualizing Your Data: Incorporating Mongo into Loggly Infrastructure
Avatar for mongodb mongodb
50
10k
Are puppies a ranking factor?
Avatar for Jono Alderson jonoalderson
2
3.9k
Agile that works and the tools we love
Avatar for Rasmus Luckow-Nielsen rasmusluckow
331
22k

Transcript

  1. -FBSOJOH
 $SPTTNPEBM&NCFEEJOHT
 GPS$PPLJOH3FDJQFT
 BOE'PPE*NBHFT 2017.08.06 KantoCV #41

  2. w -FT[FL ϨγΣοΫ  w ϙʔϥϯυग़਎ w ΫοΫύουͷݚڀ։ൃ෦ w ը૾ղੳ

    w ϙʔϥϯυͷΫοΫύου Ͱˌ̍ͷϨγϐɿ
 ๻ͷϐΤϩΪ ࣗݾ঺հ

  3. 2VJ[

  4. A D G J B E H K C I

    F

  5. A D G J B E H K C I

    F

  6. C B A 1 2 3

  7. C B A 1 2 3

  8. C B A 1 2 3

  9. C B A 1 2 3

  10. A B C D E F G

  11. A B C D E F G

  12. w $MBTTJpDBUJPOPGSFDJQFT JT)"3% w *NBHFSFDPHOJUJPOPO GPPEJT)"3% w JOUSBDMBTTWBSJFUZ w JOUFSDMBTTTJNJMBSJUZ

    w 5IFFYJTUJOHEBUBTFUT BSF4."-- w 5PPNBOZJOHSFEJFOUT w $VMUVSBMCJBTFT w &YJTUJOHTPMVUJPOTIBWF MPXBDDVSBDZ http://www.robotshop.com/

  13. July 20, 2017

  14. July 20, 2017

  15. July 20, 2017

  16. July 20, 2017

  17. July 20, 2017

  18. July 20, 2017

  19. 3FDJQF. %BUBTFU

  20. None
  21. None

  22. https://cookpad.com/recipe/1758645

  23. %BUBTFU w EJ⒎FSFOUTJUFTBTTPVSDF JODM$PPLQBE  w NSFDJQFTBOELJNBHFT w PGSFDJQFUJUMFTBSFOPUVOJRVF w

    BWFSBHFSFDJQFIBTJOHSFEJFOUT w PGJOHSFEJFOUT JOSFDJQFT w TPNFSFDJQFTBSFPOFTUFQ 
 TPNFDPOUBJOTVCSFDJQFT w TPNFGPPEUZQFTBSFWFSZQPQVMBS
 DIPDPMBUFDIJQDPPLJFT

  24. &NCFEEJOH

  25. TGEKRGURCEG KOCIGURCEG LQKPVGODGFFKPIURCEG UOQQVJKGU ITCVKPU UWUJK JCODWTIGTU */56*5*0/

  26. .PEFM0WFSWJFX

  27. w XPSEWFDPOJOHSFEJFOUT w UCTQPGPMJWFPJM w PMJWF@PJM CFMM@QFQQFS CMBDL@QFQQFS w CJEJSFDUJPOBM-45.

  28. w TLJQJOTUSVDUJPOT-45.PO-45. w SFDJQFTUBSUBOEFOENBSLFST w DPODBUFOBUFpOBMFODPEJOHXJUIUIF JOHSFEJFOUFODPEJOH

  29. w UFTUFE7((BOE3FT/FU w DMBTTJpDBUJPOˠFNCFEEJOHMBZFS w *NBHF/FUQSFUSBJOFE pOFUVOFE w USBJOFETFQBSBUFMZGSPNUFYU

  30. w NBUDIJOHQBJSTNJYFEQBJST w 

  31. w 'PPEDBUFHPSJFTHVFTTFEGSPNSFDJQFUJUMF w FNCFEEJOHˠDBUFHPSZDMBTTJpFS w JGTBNFDBUFHPSZLreg  w λ 

  32. 5PUBMMPTTGVODUJPO

  33. &YQFSJNFOUT

  34. JNBHFˠSFDJQF

  35. SFDJQFˠJNBHF

  36. None

  37. #FUUFSUIBO)VNBO

  38. %&.0 http://wednesday.csail.mit.edu: 4242/

  39. 6OJU"DUJWBUJPOT

  40. XPSEWFDNBUI http://www.samyzaf.com/ML/nlp/nlp.html

  41. XPSEWFDNBUI man - king + woman = queen cats -

    cat + dog = dogs France - Paris + England = London London - England + Italy = Rome Italy - Rome + Paris = France

  42. .BUIPO3FDJQFT

  43. .BUIPO*NBHFT

  44. $SPTTNPEBM

  45. w 3FDJQF.EBUBTFU w JNBHFˠSFDJQF w SFDJQFˠJNBHF w CFUUFSUIBOIVNBO w JOHSFEJFOUEFUFDUJPO

    w SFDJQFBSJUINFUJD http://www.robotshop.com/ ·ͱΊ

  46. Ҏ্