michaelhitzker / camera_awesome

📸 Embedding a camera experience within your own app shouldn't be that hard. A flutter plugin to integrate awesome Android / iOS camera experience.

CamerAwesome

📸 Embedding a camera experience within your own app should't be that hard.
A flutter plugin to integrate awesome Android / iOS camera experience.

This packages provides you a fully customizable camera experience that you can use within your app.
Use our awesome built in interface or customize it as you want.

---

Native features

Here's all native features that cameraAwesome provides to the flutter side.

System	Android	iOS
🔖 Ask permissions	✅	✅
🎥 Record video	✅	✅
🔈 Enable/disable audio	✅	✅
🎞 Take photos	✅	✅
🌤 Exposure level	✅	✅
📡 Broadcast live image stream	✅	✅
👁 zoom	✅	✅
📸 Device flash support	✅	✅
⌛️ Auto focus	✅	✅
📲 Live switching camera	✅	✅
😵‍💫 Camera rotation stream	✅	✅
🤐 Background auto stop	✅	✅

---

📖  Installation and usage

Add the package in your pubspec.yaml

dependencies:
    camerawesome: ^1.0.0 
    ...

Set permissions

• iOS add these on ios/Runner/Info.plist file

<key>NSCameraUsageDescription</key>
<string>Your own description</string>

<key>NSMicrophoneUsageDescription</key>
<string>To enable microphone access when recording video</string>

<key>NSLocationWhenInUseUsageDescription</key>
<string>To enable GPS location access for Exif data</string>

• Android Change the minimum SDK version to 21 (or higher) in android/app/build.gradle

minSdkVersion 21

Import the package in your Flutter app

import 'package:camerawesome/camerawesome_plugin.dart';

---

👌 Awesome build-in interface

Just use our builder.
That's all you need to create a complete camera experience within you app.

CameraAwesomeBuilder.awesome(
    saveConfig: SaveConfig.image(
        pathBuilder: _path(),
    ),
    onMediaTap: (mediaCapture) {
        OpenFile.open(mediaCapture.filePath);
    },
),

---

🎨 Creating a custom interface

Our builder provides a custom factory.
Now you have access to the builder property and can create your own camera experience.
The camera preview will be visible behind what you will provide to our builder.

Note
Only the camera preview is not customizable yet

CameraAwesomeBuilder.custom(
    saveConfig: SaveConfig.image(
        pathBuilder: _path(),
    ),
    builder: (state) {
        // create your interface here 
    },
),

See more in documentation

Working with the custom builder

Here is the definition of our builder method.

typedef CameraLayoutBuilder = Widget Function(CameraState cameraState);

The only thing you have access is the cameraState.
Depending on which state is our camera experience you will have access to some different method.

How camerAwesome states works ?

Using the state you can do anything you need without having to think about the camera flow

• On app start we are in [PreparingCameraState]
  
• Then depending on the initialCaptureMode you set you will be [PhotoCameraState] or [VideoCameraState]
  
• Starting a video will push a [VideoRecordingCameraState]
  
• Stopping the video will push back the [VideoCameraState]
  

Also if you want to use some specific function you can use the when method so you can write like this.

state.when(
    onPhotoMode: (photoState) => photoState.start(),
    onVideoMode: (videoState) => videoState.start(),
    onVideoRecordingMode: (videoState) => videoState.pause(),
);

See more in documentation

---

🔬 Analysis mode (WIP 🚧)

This is a first step into this feature as we are currently working on this.

⚠️ Only work on Android for now

Use this to achieve

• QR-Code scanning
• facial recognition
• AI object detection
• realtime video chats

You can check an example using MLKit on Android that detect faces inside the '''example''' directory.

CameraAwesomeBuilder.awesome(
    saveConfig: SaveConfig.image(
        pathBuilder: _path(),
    ),
    onImageForAnalysis: analyzeImage,
    imageAnalysisConfig: AnalysisConfig(
        outputFormat: InputAnalysisImageFormat.nv21, // choose between jpeg / nv21 / yuv_420_888
        width: 1024,
    ),
),

MLkit recommands to use nv21 format.
For machine learning you don't need full resolution images (1024 is enough and makes computation easier)

See more in documentation

---

🐽 Setting sensors settings

Through state you can access to a SensorConfig class.

Function	Comment
setZoom	changing zoom
setFlashMode	changing flash between NONE,ON,AUTO,ALWAYS
setBrightness	change brightness level manually (better to let this auto)

All of this configurations are listenable through a stream so your UI can automatically get updated according to the actual configuration.