Ask Your Question

Revision history [back]

click to hide/show revision 1
initial version

Error while loading YAML model file using opencv in android

I am using opencv for facial landmark detection in android. I am using JNI for interaction with library. I have saved "lbfmodel.yaml" file in assets and trying to load :

const char* jnamestr2 = jenv->GetStringUTFChars(jModelFileName, NULL); cv::String stdFileName2(jnamestr2); LOGD( "jModelFileName path : %s", jnamestr2); FacemarkLBF::Params params; params.model_filename = stdFileName2; params.cascade_face = stdFileName; Ptr<facemarklbf> facemark = FacemarkLBF::create(params);

But I am receiving error :

Invalid address 0x9b2d2e08 passed to free: value not allocated Fatal signal 6 (SIGABRT), code -6 in tid 27673

I have logged the filepath and it is :

jModelFileName path : /data/user/0/com.ds.openworld/app_model/lbfmodel.yaml

I don't know what is the issue because it is crashing in native part. The model file is also large. Is that an issue?

click to hide/show revision 2
None

updated 2018-12-03 04:20:39 -0600

berak gravatar image

Error while loading YAML model file using opencv in android

I am using opencv for facial landmark detection in android. I am using JNI for interaction with library. I have saved "lbfmodel.yaml" file in assets and trying to load :

const char* jnamestr2 = jenv->GetStringUTFChars(jModelFileName, NULL);
 cv::String stdFileName2(jnamestr2);
 LOGD( "jModelFileName path : %s", jnamestr2);
 FacemarkLBF::Params params;
 params.model_filename = stdFileName2;
 params.cascade_face = stdFileName;
 Ptr<facemarklbf> Ptr<FacemarkLBF> facemark = FacemarkLBF::create(params);

FacemarkLBF::create(params);

But I am receiving error :

Invalid address 0x9b2d2e08 passed to free: value not allocated Fatal signal 6 (SIGABRT), code -6 in tid 27673

I have logged the filepath and it is :

jModelFileName path : /data/user/0/com.ds.openworld/app_model/lbfmodel.yaml

I don't know what is the issue because it is crashing in native part. The model file is also large. Is that an issue?