Object Recognition using machine learning and image source from ESP32

A Flutter project that plays specific video when the app detected the corresponding object (by reading captured image from ESP32).

Getting Started (From Very Beginning)

Tutorial on deploying your Flutter development environment:

A few resources to get you started if this is your first time getting in touch with Flutter:

For help getting started with Flutter, view the

online documentation, which offers tutorials, samples, guidance on mobile development, and a full API reference.

Really Getting Started (with IntelliJ)

Create an empty project

Open IntelliJ and Create a new project in menu->File->Project..
Choose Flutter project and change [$yourFlutterPath] to your path of Flutter SDK, then press Next.
Change [$ProjectName], [$ProjectLocation], [$Description] and [$Organisation] into your project name, location of you project, description of your project and your organisation(You can just type anything if you do not have an organisation), then press Finish.

Now you have your first Flutter project created:
- The green box is your project structure.
- The blue box is your main working area.
- The red circle allows you to select your debugging device including simulated Android device, simulated iOS device (macOS only), real Android device and real iOS device (macOS only) if you plugged them in your computer.
- The orange circle allows you to run the App on your device if everything is ready for debug.

Modify the project

Things that need to be changed from the default empty project:
1. pubspec.yaml
  1. "image picker" and "video_player" packages is imported by adding 2 lines in [$ProjectName]/pubspec.yaml :
```
dependencies:
  flutter:
    sdk: flutter
  #newly added packages
  image_picker: ^0.5.0+6
  video_player: ^0.10.0+4 
  #end newly added packages
```
  2. 2 video are imported as assets:
    
    2 video files, "bunnyShort.mp4" and "bunny.mp4" are added to [$ProjectName]/assets/ , and 3 lines are added to [$ProjectName]/pubspec.yaml:
```
  assets:
    - assets/bunny.mp4
    - assets/bunnyShort.mp4
```
    - these 2 lines are added BELOW "flutter: " with level 2 of indentation.
  3. After you modify the pubspec.yaml,
2. ONLY if you faced AndroidX crashes in a Flutter app:
  1. First fixing method: Using Android Studio to migrate the App to Android X(Recommened by Flutter officially):
    - With Android Studio 3.2 and higher, you can quickly migrate an existing project to use AndroidX by selecting Refactor > Migrate to AndroidX from the menu bar.
    - If you encounter any problem using this approach, follows the detailed instruction by Flutter official. Or you can choose to use manual fixing as following.
  2. Second fixing method: Manually Migrate to Android X (Method used by me):
    1. In [$ProjectName]/android/gradle/wrapper/gradle-wrapper.properties change the line starting with distributionUrl like this:
      distributionUrl=https\://services.gradle.org/distributions/gradle-4.10.2-all.zip
    2. In android/build.gradle, replace:
      dependencies { classpath 'com.android.tools.build:gradle:3.2.1' }
      by
      dependencies { classpath 'com.android.tools.build:gradle:3.3.0' }
    3. In android/gradle.properties, append
      android.enableJetifier=true android.useAndroidX=true
    4. In android/app/build.gradle:
      
      Under android {, make sure compileSdkVersion and targetSdkVersion are at least 28.
    5. Replace all deprecated libraries with the AndroidX equivalents. For instance, if you’re using the default .gradle files make the following changes:
      
      In android/app/build.gradle, replace
      testInstrumentationRunner "android.support.test.runner.AndroidJUnitRunner"
      by
      testInstrumentationRunner "androidx.test.runner.AndroidJUnitRunner"
      Under dependencies {, replace
      androidTestImplementation 'com.android.support.test:runner:1.0.2' androidTestImplementation 'com.android.support.test.espresso:espresso-core:3.0.2'
      by
      androidTestImplementation 'androidx.test:runner:1.1.1' androidTestImplementation 'androidx.test.espresso:espresso-core:3.1.1'

Now you have every precautions settled.

Code

All code in Flutter should be placed in [$ProjectName]/lib, we have a total of 2 .dart file:

main.dart

videoRoute.dart

How to Create New File (Creating VideoRoute.dart)

Right click on the structure area's lib folder, choose New->Dart File.

Then named the name to videoRoute and press OK.

All things are settled, you can now debug this program with your device(s).

Debuging using real device

You can only debug this program with real device, because simulator does not support camera function.

Also, only Android device is supported at this stage.

Connect your Android device to your computer.
Click on the menu in the red circle in IntelliJ:
Then choose your device:
Then click on the run icon in IntelliJ next to the orange circle as shown above.
You can now see the debugging information in the green box:
You can now edit the source code and perform hot reload by clicking on the lightning button in the red circle above.

Getting realease

If you finish all debuging, you can have your mobile App released:

Goto the path of your project in shell:
Then input
```
flutter clean
flutter build apk
```
You should have your apk in [$yourPathOfProject]/build/app/outputs/apk.

wireMuffin / ObjectDetection_Flutter