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
What an ML-ful World! MLKit for Android dev.
Search
Britt Barak
October 12, 2018
Programming
0
130
What an ML-ful World! MLKit for Android dev.
Britt Barak
October 12, 2018
Tweet
Share
More Decks by Britt Barak
See All by Britt Barak
[Vonage] Introducing Conversations
brittbarak
1
130
Kids, Play Nice! Kotlin-Java Interop In Mind
brittbarak
2
450
Sharing is Caring- Getting Started with Kotlin Multiplatform
brittbarak
2
2.1k
Between JOMO and FOMO: You are reshaping communication.
brittbarak
2
1.2k
Build Apps For The Ones You Love
brittbarak
1
120
Make your app dance with MotionLayout
brittbarak
8
1.4k
Who's afraid of ML? V2 : First steps with MlKit
brittbarak
1
460
Oh, the places you'll go! Cracking Navigation on Android
brittbarak
0
480
The organic evolution - how and why we created peer mentorship program
brittbarak
0
54
Other Decks in Programming
See All in Programming
開発チーム・開発組織の設計改善スキルの向上
masuda220
PRO
20
11k
Cache Me If You Can
ryunen344
2
690
旅行プランAIエージェント開発の裏側
ippo012
2
900
Kiroの仕様駆動開発から見えてきたAIコーディングとの正しい付き合い方
clshinji
1
210
go test -json そして testing.T.Attr / Kyoto.go #63
utgwkk
3
290
MCPで実現するAIエージェント駆動のNext.jsアプリデバッグ手法
nyatinte
7
1.1k
How Android Uses Data Structures Behind The Scenes
l2hyunwoo
0
430
Processing Gem ベースの、2D レトロゲームエンジンの開発
tokujiros
2
130
AIでLINEスタンプを作ってみた
eycjur
1
230
そのAPI、誰のため? Androidライブラリ設計における利用者目線の実践テクニック
mkeeda
2
290
Ruby×iOSアプリ開発 ~共に歩んだエコシステムの物語~
temoki
0
270
今だからこそ入門する Server-Sent Events (SSE)
nearme_tech
PRO
3
170
Featured
See All Featured
Intergalactic Javascript Robots from Outer Space
tanoku
272
27k
Writing Fast Ruby
sferik
628
62k
Creating an realtime collaboration tool: Agile Flush - .NET Oxford
marcduiker
31
2.2k
The Power of CSS Pseudo Elements
geoffreycrofte
77
6k
Being A Developer After 40
akosma
90
590k
How STYLIGHT went responsive
nonsquared
100
5.8k
Six Lessons from altMBA
skipperchong
28
4k
Documentation Writing (for coders)
carmenintech
74
5k
Scaling GitHub
holman
463
140k
Statistics for Hackers
jakevdp
799
220k
Building a Scalable Design System with Sketch
lauravandoore
462
33k
The Art of Delivering Value - GDevCon NA Keynote
reverentgeek
15
1.6k
Transcript
What an ML-ful world Britt Barak
Once upon a time @BrittBarak
beta @BrittBarak
ML Capability ?! @BrittBarak
Who is afraid of Machine Learning? & First Steps With
ML-Kit @BrittBarak
Britt Barak Developer Experience, Nexmo Google Developer Expert Britt Barak
@brittBarak
None
@BrittBarak
= @BrittBarak
§ What’s the difference? @BrittBarak
…and classify? @BrittBarak
@BrittBarak
This is a strawberry @BrittBarak
This is a strawberry Red Seeds pattern Narrow top leaves
@BrittBarak Pointy at the bottom Round at the top
Strawberry Not Not Not Strawberry Strawberry Not Not Not @BrittBarak
~*~ images ~*~ @BrittBarak
@BrittBarak Vision library
Text Recognition @BrittBarak
Face Detection @BrittBarak
Barcode Scanning @BrittBarak
Image Labelling @BrittBarak
Landmark Recognition @BrittBarak
Custom Models @BrittBarak
Example @BrittBarak
@BrittBarak
@BrittBarak
Detector detector .execute(image) Result: @BrittBarak “Ben & Jerry’s pistachio ice
cream”
1. Setup Detector @BrittBarak
Local or cloud? @BrittBarak
@BrittBarak
Local •Realtime •Offline support •Security / Privacy •Bandwith •… @BrittBarak
Cloud •More accuracy & features •But more latency •Pricing @BrittBarak
1. Setup Detector @BrittBarak
Text Detector textDetector = FirebaseVision.getInstance() @BrittBarak
Text Detector textDetector = FirebaseVision.getInstance() .onDeviceTextRecognizer @BrittBarak
Text Detector textDetector = FirebaseVision.getInstance() .cloudTextRecognizer @BrittBarak
2. Process input @BrittBarak
FirebaseVisionImage •Bitmap •image Uri •Media Image •byteArray •byteBuffer @BrittBarak
image = FirebaseVisionImage.fromBitmap(bitmap) @BrittBarak Text Detector
3. Execute the model @BrittBarak
Text Detector textDetector.processImage(image) @BrittBarak
Text Detector textDetector.processImage(image) .addOnSuccessListener { } @BrittBarak
Text Detector textDetector.processImage(image) .addOnSuccessListener { firebaseVisionTexts -> processOutput(fbVisionTexts) } @BrittBarak
4. Process output @BrittBarak
firebaseVisionTexts.text @BrittBarak
someTextView.text = firebaseVisionTexts.text @BrittBarak UI
Result @BrittBarak
Result @BrittBarak
(another) Example : Labelling @BrittBarak
Detector detector .execute(image) Result: @BrittBarak ice cream pint
Vegetables Deserts Vegetables
1. Setup Detector @BrittBarak
Image Classifier imageDetector = FirebaseVision.getInstance() @BrittBarak
Image Classifier imageDetector = FirebaseVision.getInstance() .visionLabelDetector @BrittBarak
Image Classifier imageDetector = FirebaseVision.getInstance .visionCloudLabelDetector @BrittBarak
2. Process input @BrittBarak
image = FirebaseVisionImage.fromBitmap(bitmap) @BrittBarak Image Classifier
3. Execute the model @BrittBarak
Image Classifier imageDetector.detectInImage(image) @BrittBarak
Image Classifier imageDetector.detectInImage(image) .addOnSuccessListener{ } @BrittBarak
Image Classifier imageDetector.detectInImage(image) .addOnSuccessListener{ fBLabels -> processOutput(fBLabels) } @BrittBarak
4. Process output @BrittBarak
fbLabel.label fbLabel.confidence fbLabel.entityId @BrittBarak
UI for (fbLabel in labels) { s = "${fbLabel.label} :
${fbLabel.confidence}" } @BrittBarak
Result
Result
It is an ML-ful world Enjoy!
Thank you! Keep in touch!