I want to have a light model (light in terms of size in MB) so I switched the model from resnet50 to LightenedCNN_A model. I had to change the forward() method called from extract() method of DNNExtractor class from

net.forward(outputBlobs, Parameters.DEEP_LAYER);
to
net.forward(outputBlobs);

But I am facing 2 problems with these changes-

1. I am getting "No faces detected" although the white box around face is showing accurately. Thus, nothing is being saved under "unspecified".
2. When I add an identity from gallery or capture from video and when I click on "Save Identity" button, the app crashes.

Here is the stacktrace for the crash-

E/cv::error(): OpenCV(3.4.1) Error: Assertion failed (srcMat.dims == 2 && srcMat.cols == weights.cols && dstMat.rows == srcMat.rows && dstMat.cols == weights.rows && srcMat.type() == weights.type() && weights.type() == dstMat.type() && srcMat.type() == 5 && (biasMat.empty() || (biasMat.type() == srcMat.type() && biasMat.isContinuous() && (int)biasMat.total() == dstMat.cols))) in static void cv::dnn::FullyConnectedLayerImpl::FullyConnected::run(const cv::Mat&, const cv::Mat&, const cv::Mat&, cv::Mat&, const cv::dnn::experimental_dnn_v4::ActivationLayer*, int), file /build/master_pack-android/opencv/modules/dnn/src/layers/fully_connected_layer.cpp, line 154 2019-08-14 18:10:39.017 24057-24174/melonizippo.org.facerecognition E/org.opencv.dnn: dnn::forward_13() caught cv::Exception: OpenCV(3.4.1) /build/master_pack-android/opencv/modules/dnn/src/layers/fully_connected_layer.cpp:154: error: (-215) srcMat.dims == 2 && srcMat.cols == weights.cols && dstMat.rows == srcMat.rows && dstMat.cols == weights.rows && srcMat.type() == weights.type() && weights.type() == dstMat.type() && srcMat.type() == 5 && (biasMat.empty() || (biasMat.type() == srcMat.type() && biasMat.isContinuous() && (int)biasMat.total() == dstMat.cols)) in function static void cv::dnn::FullyConnectedLayerImpl::FullyConnected::run(const cv::Mat&, const cv::Mat&, const cv::Mat&, cv::Mat&, const cv::dnn::experimental_dnn_v4::ActivationLayer*, int) 2019-08-14 18:10:39.019 24057-24174/melonizippo.org.facerecognition E/AndroidRuntime: FATAL EXCEPTION: Thread-4 Process: melonizippo.org.facerecognition, PID: 24057 CvException [org.opencv.core.CvException: cv::Exception: OpenCV(3.4.1) /build/master_pack-android/opencv/modules/dnn/src/layers/fully_connected_layer.cpp:154: error: (-215) srcMat.dims == 2 && srcMat.cols == weights.cols && dstMat.rows == srcMat.rows && dstMat.cols == weights.rows && srcMat.type() == weights.type() && weights.type() == dstMat.type() && srcMat.type() == 5 && (biasMat.empty() || (biasMat.type() == srcMat.type() && biasMat.isContinuous() && (int)biasMat.total() == dstMat.cols)) in function static void cv::dnn::FullyConnectedLayerImpl::FullyConnected::run(const cv::Mat&, const cv::Mat&, const cv::Mat&, cv::Mat&, const cv::dnn::experimental_dnn_v4::ActivationLayer*, int) ] at org.opencv.dnn.Net.forward_3(Native Method) at org.opencv.dnn.Net.forward(Net.java:302) at melonizippo.org.facerecognition.deep.DNNExtractor.extract(DNNExtractor.java:40) at melonizippo.org.facerecognition.AddIdentityActivity.addImage(AddIdentityActivity.java:626) at melonizippo.org.facerecognition.AddIdentityActivity.addImage(AddIdentityActivity.java:605) at melonizippo.org.facerecognition.AddIdentityActivity.addImage(AddIdentityActivity.java:598) at melonizippo.org.facerecognition.AddIdentityActivity.lambda$processImages$10$AddIdentityActivity(AddIdentityActivity.java:533) at melonizippo.org.facerecognition.AddIdentityActivity$$Lambda$8.run(Unknown Source:20) at java.lang.Thread.run(Thread.java:764)

I am missing something? Some change I am failing to do to make this model work for face detection and saving the identity?

Hi, I am running the project as it is from android studio and getting an exception about not finding 'resnet50_ft.caffemodel'

Here is the stacktrace-

E/cv::error(): OpenCV(3.4.1) Error: Unspecified error (FAILED: fs.is_open(). Can't open "/data/user/0/melonizippo.org.facerecognition/cache/caffe_model/resnet50_ft_caffe/resnet50_ft.caffemodel") in bool cv::dnn::ReadProtoFromBinaryFile(const char*, google::protobuf::Message*), file /build/master_pack-android/opencv/modules/dnn/src/caffe/caffe_io.cpp, line 1126
2019-08-08 10:41:11.911 5421-5421/? E/org.opencv.dnn: dnn::readNetFromCaffe_10() caught cv::Exception: OpenCV(3.4.1) /build/master_pack-android/opencv/modules/dnn/src/caffe/caffe_io.cpp:1126: error: (-2) FAILED: fs.is_open(). Can't open "/data/user/0/melonizippo.org.facerecognition/cache/caffe_model/resnet50_ft_caffe/resnet50_ft.caffemodel" in function bool cv::dnn::ReadProtoFromBinaryFile(const char*, google::protobuf::Message*)
2019-08-08 10:41:11.915 5421-5421/? E/AndroidRuntime: FATAL EXCEPTION: main
    Process: melonizippo.org.facerecognition, PID: 5421
    java.lang.RuntimeException: Unable to create application melonizippo.org.facerecognition.FaceRecognitionApp: CvException [org.opencv.core.CvException: cv::Exception: OpenCV(3.4.1) /build/master_pack-android/opencv/modules/dnn/src/caffe/caffe_io.cpp:1126: error: (-2) FAILED: fs.is_open(). Can't open "/data/user/0/melonizippo.org.facerecognition/cache/caffe_model/resnet50_ft_caffe/resnet50_ft.caffemodel" in function bool cv::dnn::ReadProtoFromBinaryFile(const char*, google::protobuf::Message*)
    ]
        at android.app.ActivityThread.handleBindApplication(ActivityThread.java:6112)
        at android.app.ActivityThread.access$1200(ActivityThread.java:212)
        at android.app.ActivityThread$H.handleMessage(ActivityThread.java:1806)
        at android.os.Handler.dispatchMessage(Handler.java:106)
        at android.os.Looper.loop(Looper.java:193)
        at android.app.ActivityThread.main(ActivityThread.java:6912)
        at java.lang.reflect.Method.invoke(Native Method)
        at com.android.internal.os.RuntimeInit$MethodAndArgsCaller.run(RuntimeInit.java:493)
        at com.android.internal.os.ZygoteInit.main(ZygoteInit.java:860)
     Caused by: CvException [org.opencv.core.CvException: cv::Exception: OpenCV(3.4.1) /build/master_pack-android/opencv/modules/dnn/src/caffe/caffe_io.cpp:1126: error: (-2) FAILED: fs.is_open(). Can't open "/data/user/0/melonizippo.org.facerecognition/cache/caffe_model/resnet50_ft_caffe/resnet50_ft.caffemodel" in function bool cv::dnn::ReadProtoFromBinaryFile(const char*, google::protobuf::Message*)
    ]
        at org.opencv.dnn.Dnn.readNetFromCaffe_0(Native Method)
        at org.opencv.dnn.Dnn.readNetFromCaffe(Dnn.java:107)
        at melonizippo.org.facerecognition.deep.DNNExtractor.<init>(DNNExtractor.java:26)
        at melonizippo.org.facerecognition.FaceRecognitionApp.initFaceRecognition(FaceRecognitionApp.java:94)
        at melonizippo.org.facerecognition.FaceRecognitionApp.onCreate(FaceRecognitionApp.java:70)
        at android.app.Instrumentation.callApplicationOnCreate(Instrumentation.java:1176)
        at android.app.ActivityThread.handleBindApplication(ActivityThread.java:6107)
        	... 8 more

Do I have to train the model myself and generate the said file?