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
ML Kit Introduction (for Android)
Search
Elvis Lin
July 18, 2018
Programming
0
290
ML Kit Introduction (for Android)
Introduce the basic concept of ML Kit and how to use it in Android development
Elvis Lin
July 18, 2018
Tweet
Share
More Decks by Elvis Lin
See All by Elvis Lin
Protect Users' Privacy in iOS 14
elvismetaphor
0
49
Dubugging Tips and Tricks for iOS development
elvismetaphor
0
50
Strategies of Facebook LightSpeed project
elvismetaphor
0
79
Background Execution And WorkManager
elvismetaphor
2
480
作為一個跨平台的 Mobile App 開發者,從入門到放棄!?
elvismetaphor
2
510
Dependency Injection for testability of iOS app
elvismetaphor
1
1.4k
Briefly Introduction of Kotlin coroutines
elvismetaphor
1
280
MotionLayout Brief Introduction
elvismetaphor
1
330
Chapter 10. Pattern Matching with Regular Expressions
elvismetaphor
0
46
Other Decks in Programming
See All in Programming
Startups on Rails in Past, Present and Future–Irina Nazarova, RailsConf 2025
irinanazarova
0
280
「テストは愚直&&網羅的に書くほどよい」という誤解 / Test Smarter, Not Harder
munetoshi
1
220
バイブコーディング超えてバイブデプロイ〜CloudflareMCPで実現する、未来のアプリケーションデリバリー〜
azukiazusa1
2
670
リッチエディターを安全に開発・運用するために
unachang113
1
230
ISUCON研修おかわり会 講義スライド
arfes0e2b3c
1
480
Python型ヒント完全ガイド 初心者でも分かる、現代的で実践的な使い方
mickey_kubo
1
250
ご注文の差分はこちらですか? 〜 AWS CDK のいろいろな差分検出と安全なデプロイ
konokenj
4
640
CIを整備してメンテナンスを生成AIに任せる
hazumirr
0
140
Advanced Micro Frontends: Multi Version/ Framework Scenarios @WAD 2025, Berlin
manfredsteyer
PRO
0
420
AWS Summit Japan 2024と2025の比較/はじめてのKiro、今あなたは岐路に立つ
satoshi256kbyte
1
220
「App Intent」よくわからんけどすごい!
rinngo0302
1
120
PHPUnitの限界をPlaywrightで補完するテストアプローチ
yuzneri
0
260
Featured
See All Featured
ReactJS: Keep Simple. Everything can be a component!
pedronauck
667
120k
Easily Structure & Communicate Ideas using Wireframe
afnizarnur
194
16k
Rebuilding a faster, lazier Slack
samanthasiow
83
9.1k
Building Applications with DynamoDB
mza
95
6.5k
Dealing with People You Can't Stand - Big Design 2015
cassininazir
367
26k
Done Done
chrislema
184
16k
The World Runs on Bad Software
bkeepers
PRO
70
11k
Documentation Writing (for coders)
carmenintech
72
4.9k
Building Flexible Design Systems
yeseniaperezcruz
328
39k
The Pragmatic Product Professional
lauravandoore
35
6.8k
Site-Speed That Sticks
csswizardry
10
710
JavaScript: Past, Present, and Future - NDC Porto 2020
reverentgeek
50
5.5k
Transcript
ML Kit 使⽤用簡介 Elvis Lin @Android Taipei 2018-07-18
關於我 • Elvis Lin • Android 與 iOS 永遠的初學者 •
Twitter: @elvismetaphor • Blog: https://blog.elvismetaphor.me
不是業配 https://youtu.be/Z-dqGRSsaBs
⼤大綱 • 什什麼是(我理理解的)機器學習 • 移動裝置上實作機器學習應⽤用的限制 • TensorFlow Lite 與 ML
Kit • 範例例
機器學習的應⽤用
機器學習 • 從資料中歸納出有⽤用的規則 • 訓練模型 • 使⽤用模型 • Mobile Application
Engineer 參參與開發主要是在「使⽤用模型」 這個範圍
Data Result (Trained) Model
移動裝置上 實作機器學習應⽤用的限制 • 記憶體有限與儲存空間有限 • 計算能⼒力力不如⼤大型伺服器 • 電池容量量有限
移動裝置上 實作機器學習應⽤用的改良⽅方向 • 記憶體有限與儲存空間有限 —> 減少模型(Model)的體積 • 計算能⼒力力不如⼤大型伺服器 —> 降低演算法的複雜度
• 電池容量量有限 —> 降低演算法的複雜度
Google 推出的解決⽅方案 • TensorFlow Lite • ML Kit
https://www.tensorflow.org/mobile/tflite/
Neural Networks API Metal
ML Kit • Cloud Vision API / Mobile Vision API
• Tensorflow Lite • 整合 Firebase,託管「客製化的模型」
ML Kit Base APIs • Image labeling • Text recognition
(OCR) • Face detection • Barcode scanning • Landmark detection • others……
使⽤用 ML Kit
建立⼀一個 Firebase 專案
建立⼀一個 Android app 下載設定檔 設定好 Package Name 下載 google-service.json
<root>/build.gradle dependencies { classpath 'com.android.tools.build:gradle:3.1.3' classpath 'com.google.gms:google-services:4.0.2' }
<root>/app/build.gradle dependencies { // ... implementation 'com.google.firebase:firebase-ml-vision:16.0.0' }
掃描 barcode (local) FirebaseVisionImage image = FirebaseVisionImage.fromBitmap(image); FirebaseVisionBarcodeDetectorOptions options =
new FirebaseVisionBarcodeDetectorOptions.Builder() .setBarcodeFormats( FirebaseVisionBarcode.FORMAT_QR_CODE, FirebaseVisionBarcode.FORMAT_AZTEC ) .build(); FirebaseVisionBarcodeDetector detector = FirebaseVision.getInstance() .getVisionBarcodeDetector(options); detector.detectInImage(image) .addOnSuccessListener( new OnSuccessListener<List<FirebaseVisionBarcode>>() { @Override public void onSuccess(List<FirebaseVisionBarcode> barcodes) {} }) .addOnFailureListener(new OnFailureListener() { @Override public void onFailure(@NonNull Exception e) {} });
初始化 Detector FirebaseVisionImage image = FirebaseVisionImage.fromBitmap(image); FirebaseVisionBarcodeDetectorOptions options = new
FirebaseVisionBarcodeDetectorOptions.Builder() .setBarcodeFormats( FirebaseVisionBarcode.FORMAT_QR_CODE, FirebaseVisionBarcode.FORMAT_AZTEC ) .build(); FirebaseVisionBarcodeDetector detector = FirebaseVision .getInstance() .getVisionBarcodeDetector(options);
取得結果 detector.detectInImage(image) .addOnSuccessListener( new OnSuccessListener<List<FirebaseVisionBarcode>>() { @Override public void onSuccess(List<FirebaseVisionBarcode>
barcodes) {} }) .addOnFailureListener(new OnFailureListener() { @Override public void onFailure(@NonNull Exception e) {} });
⽀支援的 barcode 格式 • Code 128 (FORMAT_CODE_128) • Code 39
(FORMAT_CODE_39) • Code 93 (FORMAT_CODE_93) • Codabar (FORMAT_CODABAR) • EAN-13 (FORMAT_EAN_13) • EAN-8 (FORMAT_EAN_8) • ITF (FORMAT_ITF) • UPC-A (FORMAT_UPC_A) • UPC-E (FORMAT_UPC_E) •QR Code (FORMAT_QR_CODE) • PDF417 (FORMAT_PDF417) • Aztec (FORMAT_AZTEC) • Data Matrix (FORMAT_DATA_MATRIX)
辨識⽂文字 (local) FirebaseVisionImage image = FirebaseVisionImage.fromBitmap(selectedImage); FirebaseVisionTextDetector detector = FirebaseVision.getInstance().getVisionTextDetector();
detector.detectInImage(image) .addOnSuccessListener(new OnSuccessListener<FirebaseVisionText>() { @Override public void onSuccess(FirebaseVisionText text) {} }) .addOnFailureListener(new OnFailureListener() { @Override public void onFailure(@NonNull Exception e) {} });
辨識⽂文字 (cloud) FirebaseVisionCloudDetectorOptions options = new FirebaseVisionCloudDetectorOptions.Builder() .setModelType(FirebaseVisionCloudDetectorOptions.LATEST_MODEL) .setMaxResults(15) .build();
FirebaseVisionImage image = FirebaseVisionImage.fromBitmap(selectedImage); FirebaseVisionCloudDocumentTextDetector detector = FirebaseVision.getInstance() .getVisionCloudDocumentTextDetector(options); detector.detectInImage(image) .addOnSuccessListener(new OnSuccessListener<FirebaseVisionCloudText>() { @Override public void onSuccess(FirebaseVisionCloudText text) {} }) .addOnFailureListener(new OnFailureListener() { @Override public void onFailure(@NonNull Exception e) {} });
補充資料 • ML Kit 簡介 (for Android) https://blog.elvismetaphor.me/ml-kit-fundamentals-for- android-6444e2db0fdb •
ML Kit 簡介 (for iOS) https://blog.elvismetaphor.me/ml-kit-fundamentals-for- ios-cb705044e69b
參參考資料 • https://youtu.be/Z-dqGRSsaBs • https://codelabs.developers.google.com/codelabs/mlkit- android/ • https://github.com/firebase/quickstart-android/tree/ master/mlkit
None