salkuadrat / learning Goto Github PK

Can i use learning_face_detection package to fix the selfie photo orientation?

when i add this dependency google_ml_kit: ^0.7.2 it crash the project when i do the selfie segmentation.
on this below line.

SegmentationMask? mask = await _segmenter.process(image);

Build Failed in ioS showing issue Swift Compiler Error (Xcode): No such module 'MLKit'
if Anyone find the Solution Please solve it Swift Compiler Error (Xcode): No such module 'MLKit'

Can the model download address in learning_digital_ink_recognition come from other locations, such as our own server?

Data download scheduled for file: https://dl.google.com/handwriting/models/qrnn.en.reco_20200318.fst_20191208.recospec.zip
Data download scheduled for file: https://dl.google.com/handwriting/models/en.20191208.compact.fst.zip
Starting download: https://dl.google.com/handwriting/models/en.20191208.compact.fst.zip

https://dl.google.com Replace with own domain name

Hi,
how i can invoke delete offset point,
thanks

Since this package uses com.google.mlkit.vision, what about license and pricing?

Please fix the issue. As this plugin is based on very old version of camera plugin.

Hi, can I change the text 'Choose image' and 'Take photo' of the InputCameraView buttons?

Thanks for listening :)

Face detection working fine on android but doesn't work on ios (tested on both Mac M1 & Intel)

I have this problem only on iOS devices when I run the app:

" Runner[2241:317680] The selected imageFormatGroup is not supported by iOS. Defaulting to brga8888 "

Can someone help me?

I want to put a condition when it finds a specific value and sends it to another page, but it does not stop sending values ​​to the page

` Future<void> _startRecognition(InputImage image) async {
   TextRecognitionState state = Provider.of(context, listen: false);

   if (state.isNotProcessing) {
     state.startProcessing();
     state.image = image;
     state.data = await _textRecognition?.process(image);
     state.stopProcessing();
   }
   ///=======
   if(state.text.length>=12){

     Navigator.push(
       context,
       MaterialPageRoute(builder: (context) => fff(state.text)),
     );

     state.stopProcessing();
     state.removeListener(() { state;});
   }`

Hi,

First off thanks for the libraries, they're pretty awesome!

I was playing around with the digital ink recognition part of the example app and noticed that the scores of every RecognitionCandidate returned was null.

Reproduction steps

1. Add the following printout statement on line 80 of digital_ink_recognition.dart:

for (final candidate in state.data) {
  debugPrint(
      '$LearningDigitalInkRecognition: _startRecognition : text=${candidate.text} : score=${candidate.score}');
}

2. Launch the example app
3. On the main screen select "Digital Ink Recognition"
4. Draw a "W" on the canvas
5. Tap the "Start Recognition" button

Expected output

• A list of RecognitionCandidate, each with member variables text (String) and score (double?) set to non-null values when appropriate

Actual output

• A list of RecognitionCandidate which all have the variable text populated with a meaningful value but score is null
• As the log below shows, the first candidate at least should have a score as we can see it being printed out in both tf_recognizer.cc and helper.cc

Execution Log

I/flutter ( 1994): main
I/MddModelManager( 1994): isModelDownloadedAndValid()
I/MddModelManager( 1994): isModelDownloadedAndValid()
I/native  ( 1994): tf_recognizer.cc:192 inkhash: 59ce57784a4f74a8
I/native  ( 1994): tf_recognizer.cc:219 LabeledInkCurveProcessor returned 4 timesteps for 59ce57784a4f74a8.

I/native  ( 1994): tf_recognizer.cc:293 best result: 'W'  score: 52.4192

I/native  ( 1994): tf_recognizer.cc:345  1  request 17.71%    1.3ms  overall 24.61%    0.01s  LabeledInkProcessor
I/native  ( 1994):  2  request 55.93%    4.2ms  overall 51.79%    0.02s  RunTfLiteModelRunner
I/native  ( 1994):  3  request 26.36%    2.0ms  overall 23.60%    0.01s  Decoder
I/native  ( 1994):     request           7.5ms  overall           0.04s  Total
I/native  ( 1994):

I/native  ( 1994): helper.cc:160 Recognition result: W with score: 52.4192

I/flutter ( 1994): LearningDigitalInkRecognition: _startRecognition : text=W : score=null

I/flutter ( 1994): LearningDigitalInkRecognition: _startRecognition : text=w : score=null
I/flutter ( 1994): LearningDigitalInkRecognition: _startRecognition : text=Wi : score=null
I/flutter ( 1994): LearningDigitalInkRecognition: _startRecognition : text=wi : score=null
I/flutter ( 1994): LearningDigitalInkRecognition: _startRecognition : text=W  : score=null
I/flutter ( 1994): LearningDigitalInkRecognition: _startRecognition : text=WW : score=null
I/flutter ( 1994): LearningDigitalInkRecognition: _startRecognition : text=w  : score=null
I/flutter ( 1994): LearningDigitalInkRecognition: _startRecognition : text=WN : score=null
I/flutter ( 1994): LearningDigitalInkRecognition: _startRecognition : text=WV : score=null
I/flutter ( 1994): LearningDigitalInkRecognition: _startRecognition : text=Wr : score=null

Flutter Doctor

[✓] Flutter (Channel stable, 2.5.0, on macOS 11.4 20F71 darwin-arm, locale en-US)
    • Flutter version 2.5.0 at /Users/charles/NonSystemic/Code/flutter2
    • Upstream repository https://github.com/flutter/flutter.git
    • Framework revision 4cc385b4b8 (6 days ago), 2021-09-07 23:01:49 -0700
    • Engine revision f0826da7ef
    • Dart version 2.14.0

[✓] Android toolchain - develop for Android devices (Android SDK version 30.0.3)
    • Android SDK at /Users/charles/Library/Android/sdk
    • Platform android-30, build-tools 30.0.3
    • Java binary at: /Applications/Android Studio.app/Contents/jre/Contents/Home/bin/java
    • Java version OpenJDK Runtime Environment (build 11.0.10+0-b96-7281165)
    • All Android licenses accepted.

[✓] Xcode - develop for iOS and macOS
    • Xcode at /Applications/Xcode.app/Contents/Developer
    • Xcode 12.5, Build version 12E262
    • CocoaPods version 1.10.1

[✓] Android Studio (version 2020.3)
    • Android Studio at /Applications/Android Studio.app/Contents
    • Flutter plugin can be installed from:
      🔨 https://plugins.jetbrains.com/plugin/9212-flutter
    • Dart plugin can be installed from:
      🔨 https://plugins.jetbrains.com/plugin/6351-dart
    • Java version OpenJDK Runtime Environment (build 11.0.10+0-b96-7281165)

[✓] VS Code (version 1.60.0)
    • VS Code at /Applications/Visual Studio Code.app/Contents
    • Flutter extension version 3.26.0

[✓] Connected device (1 available)
    • Pixel 3a (mobile) • 95SAY0VKZU • android-arm64 • Android 11 (API 30)

• No issues found!

my app use this package worked with anydevice . But when release app and run on device android 13 app crashed. please tell me if somebody know solutions. thanks

Hi!
Great work! On android, Selfie segmentation worked really good 😄

By the way, on ios, I got this error message when selecting image.
Hope this will help.

[VERBOSE-2:ui_dart_state.cc(186)] Unhandled Exception: MissingPluginException(No implementation found for method process on channel LearningSelfieSegmentation)
#0 MethodChannel._invokeMethod
package:flutter/…/services/platform_channel.dart:156

#1 SelfieSegmenter.process
package:learning_selfie_segmentation/src/segmenter.dart:18

#2 _SelfieSegmentationPageState._process
package:bgchanger/main.dart:108

Hey @salkuadrat, thanks for the great work on this plugin - it's really helpful. I set up learning_translate and learning_language and they work perfectly on android. However, the iOS build is still failing. Not sure if the problem is related to this plugin or if it's my setup.

This is the error message when running flutter build ios. Same error happens using Xcode (and yes, I made sure to open xcworkspace and not xcodeproj as suggested here):

Swift Compiler Error (Xcode): No such module 'MLKit'
/Users/x/.pub-cache/hosted/pub.dartlang.org/learning_translate-0.0.4/ios/Classes/SwiftLearningTranslatePlugin.swift:3:7

In Xcode, it looks like this module doesn't exist either, however, there's one called "GoogleMLKit" (in case this has something to do with it (?)):

I'd greatly appreciate any help on this.

E/Vision (31182): Error loading module com.google.android.gms.vision.face optional module true: ge: No acceptable module found. Local version is 0 and remote version is 0.
I/Vision (31182): Request download for engine face is a no-op because rate limiting
I/GRALLOC (31182): LockFlexLayout: baseFormat: 11, yStride: 320, ySize: 76800, uOffset: 76800, uStride: 320
W/DynamiteModule(31182): Local module descriptor class for com.google.android.gms.vision.face not found.
I/DynamiteModule(31182): Considering local module com.google.android.gms.vision.face:0 and remote module com.google.android.gms.vision.face:0
E/Vision (31182): Error loading module com.google.android.gms.vision.face optional module true: ge: No acceptable module found. Local version is 0 and remote version is 0.
I/Vision (31182): Request download for engine face is a no-op because rate limiting
I/flutter (31182): Waiting for the face detection module to be downloaded. Please wait.
I/GRALLOC (31182): LockFlexLayout: baseFormat: 11, yStride: 320, ySize: 76800, uOffset: 76800, uStride: 320
I/DynamiteModule(31182): Considering local module com.google.android.gms.vision.dynamite:0 and remote module com.google.android.gms.vision.dynamite:2703
I/DynamiteModule(31182): Selected remote version of com.google.android.gms.vision.dynamite, version >= 2703
V/DynamiteModule(31182): Dynamite loader version >= 2, using loadModule2NoCrashUtils
I/GRALLOC (31182): LockFlexLayout: baseFormat: 11, yStride: 320, ySize: 76800, uOffset: 76800, uStride: 320
W/DynamiteModule(31182): Local module descriptor class for com.google.android.gms.vision.face not found.
I/DynamiteModule(31182): Considering local module com.google.android.gms.vision.face:0 and remote module com.google.android.gms.vision.face:0
E/Vision (31182): Error loading module com.google.android.gms.vision.face optional module true: ge: No acceptable module found. Local version is 0 and remote version is 0.
I/Vision (31182): Request download for engine face is a no-op because rate limiting
W/DynamiteModule(31182): Local module descriptor class for com.google.android.gms.vision.face not found.
I/DynamiteModule(31182): Considering local module com.google.android.gms.vision.face:0 and remote module com.google.android.gms.vision.face:0
E/Vision (31182): Error loading module com.google.android.gms.vision.face optional module true: ge: No acceptable module found. Local version is 0 and remote version is 0.
I/Vision (31182): Request download for engine face is a no-op because rate limiting
I/flutter (31182): Waiting for the face detection module to be downloaded. Please wait.
I/GRALLOC (31182): LockFlexLayout: baseFormat: 11, yStride: 320, ySize: 76800, uOffset: 76800, uStride: 320
I/DynamiteModule(31182): Considering local module com.google.android.gms.vision.dynamite:0 and remote module com.google.android.gms.vision.dynamite:2703
I/DynamiteModule(31182): Selected remote version of com.google.android.gms.vision.dynamite, version >= 2703
V/DynamiteModule(31182): Dynamite loader version >= 2, using loadModule2NoCrashUtils
I/GRALLOC (31182): LockFlexLayout: baseFormat: 11, yStride: 320, ySize: 76800, uOffset: 76800, uStride: 320

Using latest flutter and pub versions, when running I get: The 'Pods-Runner' target has transitive dependencies that include statically linked binaries: (GoogleUtilitiesComponents)

update plugin and recognize the image by image path, let the user use any camera plugin

Hello,
thanks for the lib and the compiled apk.
Unfortunately on my Samsung SM-T500 (Galaxy Tab a7) the image recongnition doesn't work using the live camera preview.
It works using camera from gallery.
I'm using learning_face_detection: ^0.0.3 and camera: ^0.9.4+5
Also I tried to add "implementation 'com.google.mlkit:face-detection:16.1.3'" in build.gradle with no success.

To be honest this seems related to this device only. On my Xiaomi phone it works. What is very strange is that on this tablet every mlkit package I tried on pub.dev doesn't work with this tablet. Could it be related to the camera plugin then? Thanks

I install and open it , it has no response to any face...too bad