qualeams / android-face-recognition-with-deep-learning-test-framework

Face Recognition framework for Android devices can be used to test different face recognition methods.

Android Face Recognition with Deep Learning - Test Framework

Continuous integration

Acknowledgements

This app was developed by Michael Sladoje and Mike Schälchli during a bachelor thesis at the Zurich University of Applied Sciences.

Acknowledgements go to the thesis supervisors Dr. Martin Loeser, Dr. Oliver Dürr, Diego Browarnik and all the contributors of our code sources.

Code has been derived from the following sources:

• OpenCV - https://github.com/opencv/opencv
• LIBSVM - https://github.com/cjlin1/libsvm
• AndroidLibSvm - https://github.com/yctung/AndroidLibSvm
• TensorFlow - https://github.com/tensorflow/tensorflow
• facenet - davidsandberg/facenet#240
• Caffe - https://github.com/BVLC/caffe
• caffe-android-demo - https://github.com/sh1r0/caffe-android-demo
• caffe-android-lib - https://github.com/sh1r0/caffe-android-lib

Library

The core functions have been moved to a separate repository which can be used as an Android library (module).

App architecture

Usage

Clone

git clone https://github.com/Qualeams/Android-Face-Recognition-with-Deep-Learning-Test-Framework.git

Compilation

APK - Package

The app can be downloaded directly from the Google Play Store - Face Recognition.

Android SDK - Java

The source can be compiled using Android Studio (common gradle scripts).

User manual

The user manual can be found here

android-face-recognition-with-deep-learning-test-framework's Issues

Build project have error:

Hi,
Error:(27, 0) Project with path ':facerecognitionlibrary/' could not be found in project ':app'.

?
Thank you!
Molys wu

Location of built tensorflow.so Library

I thinks it not correct.

Errors using 'Recognition Training' using TensorFlow

I have set classification method as TensorFlow in PreferencesHelper.java through following way

public String getClassificationMethod(){
return resources.getString(R.string.tensorflow);
}

But when I click 'Recognition Training' button in UI after setting above, I get below error

05-26 12:03:23.790 18148-18428/ch.zhaw.facerecognition E/native: tensorflow_jni.cc:269 Error during inference: Invalid argument: Session was not created with a graph before Run()!
05-26 12:10:11.683 18148-18428/ch.zhaw.facerecognition I/native: tensorflow_jni.cc:310 Image dimensions: 224x224 stride: 896
05-26 12:10:11.683 18148-18428/ch.zhaw.facerecognition I/native: tensorflow_jni.cc:233 Tensorflow: Copying Data.
05-26 12:10:11.699 18148-18428/ch.zhaw.facerecognition I/native: tensorflow_jni.cc:251 Start computing.
05-26 12:10:11.699 18148-18428/ch.zhaw.facerecognition I/native: tensorflow_jni.cc:264 End computing. Ran in 0ms (0ms avg over 2 runs)
05-26 12:10:11.699 18148-18428/ch.zhaw.facerecognition E/native: tensorflow_jni.cc:269 Error during inference: Invalid argument: Session was not created with a graph before Run()!
05-26 12:10:11.700 18148-18428/ch.zhaw.facerecognition I/System.out: 1

FATAL EXCEPTION: Thread-2410
Process: ch.zhaw.facerecognition, PID: 2209
java.lang.NumberFormatException: Invalid float: ""
at java.lang.StringToReal.invalidReal(StringToReal.java:63)
at java.lang.StringToReal.parseFloat(StringToReal.java:308)
at java.lang.Float.parseFloat(Float.java:306)
at ch.zhaw.facerecognitionlibrary.Recognition.TensorFlow.getFeatureVector(TensorFlow.java:145)
at ch.zhaw.facerecognitionlibrary.Recognition.TensorFlow.addImage(TensorFlow.java:127)
at ch.zhaw.facerecognition.Activities.TrainingActivity$1.run(TrainingActivity.java:108)
at java.lang.Thread.run(Thread.java:818)

The BPI-M3 board with Android reboots sometimes when opens the Recognition View

I have a Banana Pi board BPI-M3 with Android OS on it.
When I access Recognition View sometimes the entire system reboots. I cant even target the reason of such behavior.
I tried to debug the board with adb logcat but the latest logs don't even say that the system is going to reboot. The latest logs belong to the test app.

I'm so confused for now. What can be the reason of such behavior? Is it a hardware problem? Or maybe the system reboots when the CPU is overloaded. CPU consumption sometimes jumps up to 100% maybe this causes the board reboot?

How to detect unknown face?

can you provide any code for detecting unknown face.

How to rotate camera ?

How to rotate camera ? please help :(

Regarding Unknown Face

Hello,
The app is working fine when i train persons and then recognise them. But when move to unknown face the app display name from the previous trained persons list. I want to display unknown face when unknown face is detect.
Thanks & Regards,
Krunal Pujara

Image Reshaping with SVM doesn't work

When trying to run a "Recognition Training" for Image Reshaping with SVM, it stucks on a white screen (Training Activity) & it doesn't process any image successfully.
The same happens for an app compiled from this repo with v1.5.3 lib (imported as a module, due to an arr corrupt problem) as well as currently published version on PlayStore.

@sladomic do you have any suggestion what might go wrong please?

Also Recognition View isn't handled properly for cases when it is attempted for reco method which isn't trained yet & it crashes the app.

Best regards,
Miro

cpu run slowly

When I open the camera, Android to run slowly, I think you do not have to deal with OpenGL frame。

Thanks you for your help!

Confidence score for recognition

In the recognition view, the app gives an identity to every face inside the screen, meaning someone not added will be identified since it is printing the closest match. Is there a function in the app to extract the confidence score for recognition (how similar/different is the face to the closest match)?

Pretrained model

Hi, is there any method through which we can bring in a pre-trained model from our PCs? If yes, then what models will the app support?

How to count eye blinking while capturing image?

I want to count eye blinking while capturing images as it helps to distinguish between a real face and an image. Any way out to achieve that?

How to get recognition person name?

I want to get persons name after recognizing persons in recognition activity

Any demonstration videos for the app?

Out of Memery Error when using a trained tensorflow model to android platform

Hi, Im a senior student in China, and im trying to use a face-recognition model i've trained with tensorflow framework in android platform. But there is a problem when importing the model as shown below.
04-30 03:55:34.143 1766-1766/? E/art: Throwing OutOfMemoryError "Failed to allocate a 111118082 byte allocation with 1048576 free bytes and 94MB until OOM"
It reminds me my model is too large to import(my model's size is 108514KB which is similar to ur tensorflow model). So im eager to know how do u solve the problem. Thank u so much!

Add Person and Recognition with 'custom images/video' instead of camera

Hi

I have set of images which needs to trained for which I need to provide these as input to 'Add Person'
Currently its using camera Frames

I also have a set of images which needs to be recognized of training.
Currently its using camera Frames

Is there any way we can achieve the above task in this app, ie using custom images for training and recognition instead of using camera frames

Facilitate code coverage

Tensorflow c++ memory leak - Valgrind

I am trying to run a simple tensorflow code from one of your sources https://github.com/Qualeams/Android-Face-Recognition-with-Deep-Learning-Library/blob/17136a831d1f592026982497ef75841499de593b/jni-build/jni/tensorflow_jni.cc

I am executing simple tensorflow code to create graph def as shown below

tensorflow::NewSession (options, &session)
ReadBinaryProto (tensorflow::Env::Default(), "/home/ashok/eclipseWorkspace/faceRecognition-x86_64/Data/models/optimized_facenet.pb", &graph_def));
session->Create (graph_def);

But when I run Valgrind as shown below

valgrind --leak-check=full --show-leak-kinds=all --vex-guest-max-insns=25 ./faceRecognition-x86_64 -r -i

I get below errors

==12366== 16,000 bytes in 1 blocks are still reachable in loss record 47,782 of 47,905
==12366==    at 0x4C2E19F: operator new(unsigned long) (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==12366==    by 0xBF875DC: std::vector<tensorflow::CostModel::MemUsage, std::allocator<tensorflow::CostModel::MemUsage> >::reserve(unsigned long) (in /usr/lib/libtensorflow_cc.so)
==12366==    by 0xBF90128: tensorflow::CostModel::InitFromGraph(tensorflow::Graph const&) (in /usr/lib/libtensorflow_cc.so)
==12366==    by 0xBEE48D3: tensorflow::SimpleGraphExecutionState::InitBaseGraph(tensorflow::BuildGraphOptions const&) (in /usr/lib/libtensorflow_cc.so)
==12366==    by 0xBEE52CF: tensorflow::SimpleGraphExecutionState::MakeForBaseGraph(tensorflow::GraphDef*, tensorflow::SimpleGraphExecutionStateOptions const&, std::unique_ptr<tensorflow::SimpleGraphExecutionState, std::default_delete<tensorflow::SimpleGraphExecutionState> >*) (in /usr/lib/libtensorflow_cc.so)
==12366==    by 0xBE68B9D: tensorflow::DirectSession::MaybeInitializeExecutionState(tensorflow::GraphDef const&, bool*) (in /usr/lib/libtensorflow_cc.so)
==12366==    by 0xBE68CF9: tensorflow::DirectSession::ExtendLocked(tensorflow::GraphDef const&) (in /usr/lib/libtensorflow_cc.so)
==12366==    by 0xBE68FC7: tensorflow::DirectSession::Create(tensorflow::GraphDef const&) (in /usr/lib/libtensorflow_cc.so)
==12366==    by 0x26B899: TensorFlow::initializeRecognition() (in /home/ashok/eclipseWorkspace/faceRecognition-x86_64/Debug/faceRecognition-x86_64)
==12366==    by 0x24197D: RecognitionWithImages::RecognitionWithImages() (in /home/ashok/eclipseWorkspace/faceRecognition-x86_64/Debug/faceRecognition-x86_64)
==12366==    by 0x12F27C: main (in /home/ashok/eclipseWorkspace/faceRecognition-x86_64/Debug/faceRecognition-x86_64)

These type of errors are also generated when I do session -> run ()

Due to the above issues, the memory needed to run the program keeps increasing as time passes and the application crashes due to insufficient memory after a certain point of time.

On training

There is a requirement to shift the training function to the core java server side. Does it is required to create a Tensor Flow graph for creating training_model file using SVM?

Location of Opencv library file

I have come across a file 'CascadeClassifier' in opencv at OpenCVLibrary320/org.opencv/objdetect/CascadeClassifier.java where certain native functions appear red

Can I please know from which library are the opencv implementations of these functions present and where is the location of this library file in the project

AS , when I try to run entire project from scratch, I get below error at runtime

java.lang.UnsatisfiedLinkError: No implementation found for long org.opencv.objdetect.CascadeClassifier.CascadeClassifier_0(java.lang.String) (tried Java_org_opencv_objdetect_CascadeClassifier_CascadeClassifier_10 and Java_org_opencv_objdetect_CascadeClassifier_CascadeClassifier_10__Ljava_lang_String_2)
at org.opencv.objdetect.CascadeClassifier.CascadeClassifier_0(Native Method)
at org.opencv.objdetect.CascadeClassifier.(CascadeClassifier.java:39)
at com.example.h242018.facerecognition.Helpers.FaceDetection.setCascadeClassifier(FaceDetection.java:68)
at com.example.h242018.facerecognition.Helpers.FaceDetection.(FaceDetection.java:52)
at com.example.h242018.facerecognition.PreProcessor.PreProcessorFactory.(PreProcessorFactory.java:56)
at com.example.h242018.facerecognition.Activities.TrainingActivity$1.run(TrainingActivity.java:70)
at java.lang.Thread.run(Thread.java:818)

Cant test recognition image

Hello i just recentrly dowloaded your demo to test but after i add a new person i cant do anything with it because de buttons are not enabled ...how do i get to enable the whole app?

rebuild for error:

Hi,

Thanks anyways

molys wu

how implement eye blink detection in this source code

Build Issues

Hi , I have setup the project in Android Studio, but when I build it , I get below errors

Error:error: '-Wno-error=unused-function': No such file or directory
Error:error: '-std=c++11': No such file or directory
Error:error: '-std=c++11': No such file or directory
Error:error: '-DMIN_LOG_LEVEL=0': No such file or directory
Information:BUILD FAILED

Please help in resolving it

Eigenfaces with NN doesn't work

When trying to run a "Recognition Training" for Eigenfaces with NN, Training Activity always stucks on the last trained image, trying to save the results, exhausts all the available memory & crashes.

The same happens for an app compiled from this repo with v1.5.3 lib (imported as a module, due to an arr corrupt problem) as well as currently published version on PlayStore.

I hope the following stacktrace will help:
AndroidRuntime: FATAL EXCEPTION: Thread-4914
Process: ch.zhaw.facerecognition, PID: 18790
Theme: themes:{com.android.settings=overlay:system, default=overlay:system, iconPack:system, fontPkg:system, com.android.systemui=overlay:system, com.android.systemui.navbar=overlay:system}
java.lang.OutOfMemoryError: Failed to allocate a 85925328 byte allocation with 16777216 free bytes and 77MB until OOM
at java.lang.AbstractStringBuilder.enlargeBuffer(AbstractStringBuilder.java:95)
at java.lang.AbstractStringBuilder.append0(AbstractStringBuilder.java:146)
at java.lang.StringBuilder.append(StringBuilder.java:216)
at ch.zhaw.facerecognitionlibrary.Helpers.MatXml.dataStringBuilder(MatXml.java:283)
at ch.zhaw.facerecognitionlibrary.Helpers.MatXml.writeMat(MatXml.java:258)
at ch.zhaw.facerecognitionlibrary.Helpers.FileHelper.saveMatListToXml(FileHelper.java:129)
at ch.zhaw.facerecognitionlibrary.Recognition.Eigenfaces.saveToFile(Eigenfaces.java:132)
at ch.zhaw.facerecognitionlibrary.Recognition.Eigenfaces.train(Eigenfaces.java:68)
at ch.zhaw.facerecognition.Activities.TrainingActivity$1.run(TrainingActivity.java:143)
at java.lang.Thread.run(Thread.java:818)

Not recognizing (Nexus 5x 7.1.1)

onCameraFrame,
ppF.getProcessedImage(img, PreProcessorFactory.PreprocessingMode.RECOGNITION);
returns null.
Seems it can't find the eyes.
Using the default settings, have 2 trained classes.

Could not create Tensorflow Graph: Invalid argument: No OpKernel was registered to support Op 'Const' with these attrs

I have built Tensoflow using bazel.
I have taken the libraries from tensorflow and used in android ndk, by stetting abi as either of 'arm64-v8a' or 'armeabiv-7a'. Not Both specified together
tensorflow::Status s = session->Create(tensorflow_graph);

Below are the debug messages

06-03 15:51:03.191 25365-26625/ch.zhaw.facerecognition I/native: tensorflow_jni.cc:113 Loading Tensorflow.
06-03 15:51:03.191 25365-26625/ch.zhaw.facerecognition I/native: tensorflow_jni.cc:115 Making new SessionOptions.
06-03 15:51:03.191 25365-26625/ch.zhaw.facerecognition I/native: tensorflow_jni.cc:118 Got config, 0 devices
06-03 15:51:03.193 25365-26625/ch.zhaw.facerecognition I/native: tensorflow_jni.cc:121 Session created.
06-03 15:51:03.193 25365-26625/ch.zhaw.facerecognition I/native: tensorflow_jni.cc:124 Graph created.
06-03 15:51:03.193 25365-26625/ch.zhaw.facerecognition I/native: tensorflow_jni.cc:128 Acquired AssetManager.
06-03 15:51:03.193 25365-26625/ch.zhaw.facerecognition I/native: tensorflow_jni.cc:130 Reading file to proto: /storage/emulated/0/Pictures/facerecognition/data/TensorFlow/vgg_faces.pb
06-03 15:51:05.045 25365-26625/ch.zhaw.facerecognition I/native: tensorflow_jni.cc:135 Creating session.
06-03 15:51:05.068 25365-26625/ch.zhaw.facerecognition E/native: tensorflow_jni.cc:138 Could not create Tensorflow Graph: Invalid argument: No OpKernel was registered to support Op 'Const' with these attrs. Registered devices: [CPU], Registered kernels:

Getting compile time error.

Error:com.android.builder.internal.aapt.AaptException: Failed to crunch file C:\Users\akundu\AndroidStudioProjects\Android-Face-Recognition-with-Deep-Learning-Test-Framework-master\app\build\intermediates\exploded-aar\com.android.support\appcompat-v7\23.3.0\res\drawable-ldrtl-hdpi-v17\abc_ic_ab_back_mtrl_am_alpha.png into C:\Users\akundu\AndroidStudioProjects\Android-Face-Recognition-with-Deep-Learning-Test-Framework-master\app\build\intermediates\res\merged\debug\drawable-ldrtl-hdpi-v17\abc_ic_ab_back_mtrl_am_alpha.png

JNI functions not detected using latest gradle version

I have setup 'Android-Face-Recognition-with-Deep-Learning-Test-Framework' in Android Studio (latest version) with latest gradle version

The project builds fine.
But when I open files under 'Recognition' folder under 'facerecognitionlibrary',for eg TensorFlow.java,

native functions like
initializeTensorflow, classifyImageBmp, classifyImageRgb are not resolved and appear red as shown in attached pic.

With error message like 'Cannot resolve corresponding JNI function Java_ch_zhaw_facerecognitionlibrary_Recognition_TensorFlow_initializeTensorflow' for initializeTensorflow method

I think the latest gradle versions make use of cmakelists.txt , instead of android.mk or applications.mk
Could I please know the contents of cmakelists.txt according to latest version of gradle

Face liveness detection

While Trying to access with mobile photos , App recognition was granted. How do i make changes to not allow the mobile photos for accessing the application.

Figuring out input and output layer in a model

Hey, I would like to know how did you figure out the input and output layers for the inception 5h model as you have mentioned here.

I tried running a tool called summarize_graph in tensorflow (retrieved from here ) but I found input and output layers for inception 5h as input and output respectively.

Here's the complete output:

**Found 1 possible inputs: (name=input, type=float(1), shape=[])** 
No variables spotted.
**Found 3 possible outputs: (name=output, op=Identity) (name=output1, op=Identity) (name=output2, op=Identity)** 
Found 13462015 (13.46M) const parameters, 0 (0) variable parameters, and 0 control_edges
370 nodes assigned to device '/cpu:0'Op types used: 142 Const, 64 BiasAdd, 61 Relu, 59 Conv2D, 13 MaxPool, 9 Concat, 5 Reshape, 5 MatMul, 3 Softmax, 3 Identity, 3 AvgPool, 2 LRN, 1 Placeholder
To use with tensorflow/tools/benchmark:benchmark_model try these arguments:
bazel run tensorflow/tools/benchmark:benchmark_model -- --graph=/Users/sparker0i/Face-Recognition-1/app/src/main/assets/tensorflow_inception_graph.pb --show_flops --input_layer=input --input_layer_type=float --input_layer_shape= --output_layer=output,output1,output2

Desktop Verison

Hi,

Loved your project and would like to use it for another. Do you have a desktop version or something that can be run on a server for scalability? Preferably python, if not a problem. Would really appreciate it.

not getting this folder while cloning facerecognitionlibrary

@sladomic Sorry for this nonsense question, basically I'm new to android.

I was trying to clone this project but after complete the cloning I didn't get this folder please help me how can I get this in cloning. Folder Name is: facerecognitionlibrary

file was generated by an older version of protoc

Hi
I tried to build the project, but while building i get below error in genfiles\tensorflow\core\framework\step_stats.pb.h .

#if 3002000 < GOOGLE_PROTOBUF_MIN_PROTOC_VERSION
#error This file was generated by an older version of protoc which is
#error incompatible with your Protocol Buffer headers. Please
#error regenerate this file with a newer version of protoc.
#endif

Please help resolving it. My GOOGLE_PROTOBUF_MIN_PROTOC_VERSION is defined as
#define GOOGLE_PROTOBUF_MIN_PROTOC_VERSION 3003000

in include/google/protobuf/stubs/common.h which I have downloaded from https://github.com/google/protobuf/tree/master/src/google/protobuf as
https://github.com/Qualeams/Android-Face-Recognition-with-Deep-Learning-Library/tree/49125122ca1db5720b8be2d8cefbafdbc3ec4010/jni-build/jni/include/google/protobuf didn't had folders or required files

can not Recognise image

i can not Recognise image when i save image to database and then i get that image and match with current captured image.

en:

Delete trained data and sample images

I took 5 photos while training but I'm required to have one.

Clearing the data cache from app settings and uninstalling plus reinstalling the app keep all trained data.

Don't it have a button for this? Where is this stored?

Printing TensorFlow Library Logs in Android Studio

I have built Tensor Flow using Bazel and am using the generated TensorFlow Libraries in Android Studio
for ndk.

But I do not find the logs of Tensor Flow being printed in Android Studio.
Can I know how do I acheive this .

I have tried using Android/log.h in TensorFlow code but of no use

Runing code

Sir, please give me a video how to start the code of Android-Face-Recognition-with-Deep-Learning-Test-Framework step by step. We are use this app in our FYP but we are not understand, how to run this app code in android studio. Please help me my id is [email protected] I am very thank full to you.

Error:Configuration with name 'default' not found.

Error:Configuration with name 'default' not found.

Broken link to "VGG Face Descriptor model"

Link to "VGG Face Descriptor model"
https://www.dropbox.com/s/51wi2la5e034wfv/vgg_faces.pb?dl=0
from User Manual doesn't work anymore.

Unable to debug ndk code using breakpoints in Android Studio

Hi
I tried all possible ways to try debugging ndk code using Breakpoints, but coudn't acheive it.
Could I please know if you were able to hit breakpoints in ndk code, if so please tell any additional settings you have made

Thanks

Error:Failed to resolve: org.tensorflow:tensorflow-android:1.2.0-rc0

Error:Failed to resolve: org.tensorflow:tensorflow-android:1.2.0-rc0
Open File
Show in Project Structure dialog

How can it run on the x86 CPU

I have the device x86 CPU.
And here is my Error,that I have.
and How can I do to fix it?

E/OpenCV/StaticHelper: OpenCV error: Cannot load info library for OpenCV
D/OpenCV/StaticHelper: Library list: ""
First attempt to load libs
Trying to init OpenCV libs
D/OpenCV/StaticHelper: Trying to load library opencv_java3
D/OpenCV/StaticHelper: Cannot load library "opencv_java3"
W/System.err: java.lang.UnsatisfiedLinkError: dalvik.system.PathClassLoader[DexPathList[[zip file "/data/app/ch.zhaw.facerecognition-nZXJudycSK0TBpgCx4Pegw==/base.apk"],nativeLibraryDirectories=[/data/app/ch.zhaw.facerecognition-nZXJudycSK0TBpgCx4Pegw==/lib/x86, /system/fake-libs, /data/app/ch.zhaw.facerecognition-nZXJudycSK0TBpgCx4Pegw==/base.apk!/lib/x86, /system/lib, /vendor/lib]]] couldn't find "libopencv_java3.so"
W/System.err: at java.lang.Runtime.loadLibrary0(Runtime.java:1011)
at java.lang.System.loadLibrary(System.java:1657)
at org.opencv.android.StaticHelper.loadLibrary(StaticHelper.java:64)
at org.opencv.android.StaticHelper.initOpenCVLibs(StaticHelper.java:95)
at org.opencv.android.StaticHelper.initOpenCV(StaticHelper.java:39)
at org.opencv.android.OpenCVLoader.initDebug(OpenCVLoader.java:91)
W/System.err: at ch.zhaw.facerecognition.Activities.AddPersonPreviewActivity.(AddPersonPreviewActivity.java:66)
at java.lang.Class.newInstance(Native Method)
at android.app.Instrumentation.newActivity(Instrumentation.java:1174)
at android.app.ActivityThread.performLaunchActivity(ActivityThread.java:2669)
W/System.err: at android.app.ActivityThread.handleLaunchActivity(ActivityThread.java:2856)
at android.app.ActivityThread.-wrap11(Unknown Source:0)
at android.app.ActivityThread$H.handleMessage(ActivityThread.java:1589)
at android.os.Handler.dispatchMessage(Handler.java:106)
at android.os.Looper.loop(Looper.java:164)
W/System.err: at android.app.ActivityThread.main(ActivityThread.java:6494)
at java.lang.reflect.Method.invoke(Native Method)
at com.android.internal.os.RuntimeInit$MethodAndArgsCaller.run(RuntimeInit.java:438)
at com.android.internal.os.ZygoteInit.main(ZygoteInit.java:807)
D/OpenCV/StaticHelper: First attempt to load libs fails
D/CameraBridge: Attr count: 6
E/zygote: No implementation found for long org.opencv.objdetect.CascadeClassifier.CascadeClassifier_0(java.lang.String) (tried Java_org_opencv_objdetect_CascadeClassifier_CascadeClassifier_10 and Java_org_opencv_objdetect_CascadeClassifier_CascadeClassifier_10__Ljava_lang_String_2)
D/AndroidRuntime: Shutting down VM
E/AndroidRuntime: FATAL EXCEPTION: main
Process: ch.zhaw.facerecognition, PID: 4140
java.lang.UnsatisfiedLinkError: No implementation found for long org.opencv.objdetect.CascadeClassifier.CascadeClassifier_0(java.lang.String) (tried Java_org_opencv_objdetect_CascadeClassifier_CascadeClassifier_10 and Java_org_opencv_objdetect_CascadeClassifier_CascadeClassifier_10__Ljava_lang_String_2)
at org.opencv.objdetect.CascadeClassifier.CascadeClassifier_0(Native Method)
at org.opencv.objdetect.CascadeClassifier.(CascadeClassifier.java:31)
at ch.zhaw.facerecognitionlibrary.Helpers.FaceDetection.setCascadeClassifier(FaceDetection.java:68)
at ch.zhaw.facerecognitionlibrary.Helpers.FaceDetection.(FaceDetection.java:52)
at ch.zhaw.facerecognitionlibrary.PreProcessor.PreProcessorFactory.(PreProcessorFactory.java:56)
at ch.zhaw.facerecognition.Activities.AddPersonPreviewActivity.onResume(AddPersonPreviewActivity.java:207)
at android.app.Instrumentation.callActivityOnResume(Instrumentation.java:1355)
at android.app.Activity.performResume(Activity.java:7117)
at android.app.ActivityThread.performResumeActivity(ActivityThread.java:3556)
at android.app.ActivityThread.handleResumeActivity(ActivityThread.java:3621)
at android.app.ActivityThread.handleLaunchActivity(ActivityThread.java:2862)
at android.app.ActivityThread.-wrap11(Unknown Source:0)
at android.app.ActivityThread$H.handleMessage(ActivityThread.java:1589)
at android.os.Handler.dispatchMessage(Handler.java:106)
at android.os.Looper.loop(Looper.java:164)
at android.app.ActivityThread.main(ActivityThread.java:6494)
at java.lang.reflect.Method.invoke(Native Method)
at com.android.internal.os.RuntimeInit$MethodAndArgsCaller.run(RuntimeInit.java:438)
at com.android.internal.os.ZygoteInit.main(ZygoteInit.java:807)

Information on Size

Hello,
This is not an issue.
I wanted to ask that how did you manage to have a 17 Mb APK for this?
Please share about this.
Thank you

How to change the default settings?

I am now trying to run this on Android. I can not find the file to change the default settings when I change the model. Can anyone kindly tell me? Thanks!

I get the warning of Configuration

Configuration 'provided' is obsolete and has been replaced with 'compileOnly'.
It will be removed at the end of 2018. For more information see: http://d.android.com/r/tools/update-dependency-configurations.html

I have used implementation instead of compile and provided.

On accuracy evaluation

As I know F1 score is usually used for accuracy evaluation. But I don't know the following concepts,

In TestActivity.java,
double accuracy_reference = (double) matches_reference / (double) total_reference;
double accuracy_deviation = (double) matches_deviation / (double) total_deviation;
double robustness = accuracy_deviation / accuracy_reference;

Could you give out explanations on these concepts?

Could not find method compile()

Error:(32, 0) Could not find method compile() for arguments [project ':app'] on object of type org.gradle.api.internal.artifacts.dsl.dependencies.DefaultDependencyHandler.
Open File