Android-Object-Detection
Android-Object-Detection copied to clipboard
tzutalin
:coffee: Fast-RCNN and Scene Recognition using Caffe
download_url = 'https://www.dropbox.com/s/0i2fr9krb8wv8mp/phone_data.tar?dl=1' this path is not exists,how do I get these models.
what has to be stored in this location, could u specify it There should be model and weight in /sdcard/fastrcnn and /sdcard/vision_scene.? Thank u
i have read [python demo](https://github.com/rbgirshick/fast-rcnn/blob/master/tools/demo.py),it doesn't need DETECT_MEAN_FILE、DETECT_SYNSET_FILE,so where to get this two file thanks!
How can I build the JNI libraries from source codes?Could you please publish the source codes of libobjrek.so and libobjrek_jni.so? Thanks in advance!
crash and show this error ``` java.lang.VerifyError: com/tzutalin/vision/demo/Camera2BasicFragment at com.tzutalin.vision.demo.CameraActivity.onCreate(CameraActivity.java:55) ``` in this code in (.replace(R.id.container, Camera2BasicFragment.newInstance())) ``` if (avialbe_permission && null == savedInstanceState) { getFragmentManager().beginTransaction() .replace(R.id.container, Camera2BasicFragment.newInstance()) .commit(); }...
Is it possible to use the faster-rcnn model rather than the fast-rcnn one? Anyone has insights on what would need to change?
How can I build the JNI libraries from source codes? It seems this repository does not have source codes of JNI libraries?
Dear developers, how complex would be for you to prepare this source code as a fully functional bundle, which would serve as object recognition for visually impaired Android users. So...
08-16 13:23:36.059 29032-29065/com.tzutalin.vision.visionrecognition D/OpenGLRenderer: Use EGL_SWAP_BEHAVIOR_PRESERVED: true 08-16 13:23:36.096 29032-29065/com.tzutalin.vision.visionrecognition I/Adreno: QUALCOMM build : 63c06b2, I8366cd0437 Build Date : 12/06/15 OpenGL ES Shader Compiler Version: XE031.05.13.02 Local Branch : mybranch17112971...
