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
YGLF_2019.pdf
Search
Asim Hussain
April 05, 2019
Technology
1
150
YGLF_2019.pdf
Asim Hussain
April 05, 2019
Tweet
Share
More Decks by Asim Hussain
See All by Asim Hussain
JavaScript Saves The World - DotJS 2019
jawache
0
900
Saving the world, one line at a time [CodeLeaders Australia 2019]
jawache
0
81
AI + JavaScript Rocks @ GDGDevFest UA 2018
jawache
0
54
How to hack a web app? WebConfAsia 2018
jawache
0
94
How to scale an SPA? @ AmsterdamJS 2018
jawache
0
33
How to hack an Angular app? - ngConf 2018
jawache
0
980
Getting started with node.js @ AngleBrackets 2018
jawache
1
180
How to hack an Angular app? @ ngVikings 2018
jawache
1
1.1k
How to hack a python app? @ PyCaribbean 2018
jawache
0
180
Other Decks in Technology
See All in Technology
20251027_findyさん_音声エージェントLT
almondo_event
2
490
AIの個性を理解し、指揮する
shoota
3
450
組織全員で向き合うAI Readyなデータ利活用
gappy50
5
1.6k
Okta Identity Governanceで実現する最小権限の原則
demaecan
0
200
アウトプットから始めるOSSコントリビューション 〜eslint-plugin-vueの場合〜 #vuefes
bengo4com
3
1.8k
JSConf JPのwebsiteをGatsbyからNext.jsに移行した話 - Next.jsの多言語静的サイトと課題
leko
2
190
知覚とデザイン
rinchoku
1
630
AIプロダクトのプロンプト実践テクニック / Practical Techniques for AI Product Prompts
saka2jp
0
120
Open Table Format (OTF) が必要になった背景とその機能 (2025.10.28)
simosako
2
430
RemoteFunctionを使ったコロケーション
mkazutaka
1
140
【SORACOM UG Explorer 2025】さらなる10年へ ~ SORACOM MVC 発表
soracom
PRO
0
170
AI機能プロジェクト炎上の 3つのしくじりと学び
nakawai
0
150
Featured
See All Featured
Speed Design
sergeychernyshev
32
1.2k
Responsive Adventures: Dirty Tricks From The Dark Corners of Front-End
smashingmag
253
22k
Designing Dashboards & Data Visualisations in Web Apps
destraynor
231
53k
Visualization
eitanlees
150
16k
XXLCSS - How to scale CSS and keep your sanity
sugarenia
249
1.3M
Save Time (by Creating Custom Rails Generators)
garrettdimon
PRO
32
1.7k
Building Better People: How to give real-time feedback that sticks.
wjessup
370
20k
Designing Experiences People Love
moore
142
24k
Put a Button on it: Removing Barriers to Going Fast.
kastner
60
4k
Why You Should Never Use an ORM
jnunemaker
PRO
59
9.6k
ReactJS: Keep Simple. Everything can be a component!
pedronauck
667
130k
Imperfection Machines: The Place of Print at Facebook
scottboms
269
13k
Transcript
The Future of Machine Learning & JavaScript @jawache YGLF 2019
Asim Hussain @jawache codecraft.tv microsoft.com
https://aka.ms/jawache-cda @jawache
@jawache https://www.palinternship.com/
@jawache
Asim Web Development Machine Learning This is @EleanorHaproff's slide
None
@jawache
TheMojifier™ @jawache
None
@jawache themojifier.com
None
How to Calculate Emotion? @jawache
(1) Detect Facial Features @jawache
https://towardsdatascience.com/facial-keypoints-detection-deep-learning-737547f73515
(2) Use a Neural Network @jawache
Neural Networks Axon Dendrites Axons Body @jawache
1 23 8.6 -0.5 2.1 Activation Function @jawache Neural Networks
1 23 8.6 -0.5 2.1 x x activation(...) = -11.5
= 18.06 7.01 !-> !-> } @jawache Neural Networks
Output 0 0 1 Input @jawache Neural Networks
1.1 4.2 0.3 4 12 93 3 @jawache Neural Networks
1.1 4.2 0.3 4 12 93 @jawache 8 - 8
= -5 3 Neural Networks
1.1 4.2 0.3 4 12 93 @jawache - 8 =
-5 3 8 Neural Networks
0.1 9.2 0.2 4 12 93 @jawache 8 8 Neural
Networks
@jawache https://azure.microsoft.com/services/cognitive-services/face/
https:!//<region>.api.cognitive.microsoft.com/face/v1.0/detect { "url": "<path-to-image>" } @jawache
@jawache
Summary @jawache
• Neural Networks are incredibly powerful • Conceptually, they are
simple to understand @jawache Summary
TensorFlow, MobileNet & I'm fine @jawache
@jawache
@jawache
@jawache
TensorFlow.js @jawache
TensorFlow.js Train models Load pre-trained models @jawache
https://github.com/tensorflow/tfjs-models @jawache MobileNet
https://azure.microsoft.com/services/cognitive-services/computer-vision/ @jawache
https://codepen.io/sdras/full/jawPGa/ @jawache
@jawache https://twitter.com/ollee/status/930303340516216832
@jawache https://twitter.com/FrontendNE/status/930120267992616960
@jawache https://twitter.com/chrispiecom/status/930407801402347520
Summary @jawache
• TensorFlow.js doesn't have any dependancies • MobileNet is a
simple way to analyse images • Azure Computer Vision API ❤ @jawache Summary
Image2Image @jawache
DEMO @jawache https://zaidalyafeai.github.io/pix2pix/cats.html
@jawache Generator Discriminator ✅ ❌
@jawache Generator Discriminator ✅ ❌
@jawache Generator Discriminator ✅ ✅
@jawache
@jawache
@jawache
@jawache https://github.com/NVIDIA/vid2vid
@jawache https://github.com/NVIDIA/vid2vid
@jawache https://github.com/NVIDIA/vid2vid
@jawache https://github.com/hanzhanggit/StackGAN
Summary @jawache
• GANs learn to generate new images • They take
a lot of compute to train • But the generator model can be run in the browser @jawache Summary
@jawache aka.ms/mojifier
Asim Hussain @jawache codecraft.tv microsoft.com