如何增加一个告知FPS的函数，实时显示再输出视频图像上用于查看帧率

pi@raspberrypi:~/Documents/Paddle-Lite-Demo-master/PaddleLite-armlinux-demo/image_classification_demo $ sudo ./run.sh 
-- The C compiler identification is GNU 8.3.0
-- The CXX compiler identification is GNU 8.3.0
-- Check for working C compiler: /usr/bin/arm-linux-gnueabihf-gcc
-- Check for working C compiler: /usr/bin/arm-linux-gnueabihf-gcc -- works
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Detecting C compile features
-- Detecting C compile features - done
-- Check for working CXX compiler: /usr/bin/arm-linux-gnueabihf-g++
-- Check for working CXX compiler: /usr/bin/arm-linux-gnueabihf-g++ -- works
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- TARGET ARCH ABI: armv7hf
-- PADDLE LITE DIR: ../Paddle-Lite
-- Found OpenMP_C: -fopenmp (found version "4.5") 
-- Found OpenMP_CXX: -fopenmp (found version "4.5") 
-- Found OpenMP: TRUE (found version "4.5")  
-- Found OpenMP  4.5
-- OpenMP C flags:  -fopenmp
-- OpenMP CXX flags:  -fopenmp
-- OpenMP OpenMP_CXX_LIB_NAMES:  gomp;pthread
-- OpenMP OpenMP_CXX_LIBRARIES:  /usr/lib/gcc/arm-linux-gnueabihf/8/libgomp.so;/usr/lib/arm-linux-gnueabihf/libpthread.so
CMake Error at CMakeLists.txt:42 (find_package):
  By not providing "FindOpenCV.cmake" in CMAKE_MODULE_PATH this project has
  asked CMake to find a package configuration file provided by "OpenCV", but
  CMake did not find one.

  Could not find a package configuration file provided by "OpenCV" with any
  of the following names:

    OpenCVConfig.cmake
    opencv-config.cmake

  Add the installation prefix of "OpenCV" to CMAKE_PREFIX_PATH or set
  "OpenCV_DIR" to a directory containing one of the above files.  If "OpenCV"
  provides a separate development package or SDK, be sure it has been
  installed.


-- Configuring incomplete, errors occurred!
See also "/home/pi/Documents/Paddle-Lite-Demo-master/PaddleLite-armlinux-demo/image_classification_demo/build/CMakeFiles/CMakeOutput.log".
make: *** 没有指明目标并且找不到 makefile。 停止。
./run.sh:行19: ./image_classification_demo: 没有那个文件或目录

采用的是已经编译好的
opencv_contrib_python-4.1.0.25-cp37-cp37m-linux_armv7l.whl
opencv_python-4.4.0.46-cp37-cp37m-linux_armv7l.whl
+
python3.7.3，直接在树莓派上pip install 的

已经可以在python3.7中 Import cv2

是一定要在树莓派本机自行编译opencv+opencv_contrib么？并且在本机自行尝试cmake的时候提示

cmake fatal error: OpenCL_INCLUDE_DIR-NOTFOUND/CL/cl.h

请问如何解决？？？

....
[100%] Linking CXX executable image_classification_demo
/usr/bin/ld: //usr/lib/arm-linux-gnueabihf/libavcodec.so.58: undefined reference to `bcm_host_is_fkms_active'
collect2: error: ld returned 1 exit status
make[2]: *** [CMakeFiles/image_classification_demo.dir/build.make:136：image_classification_demo] 错误 1
make[1]: *** [CMakeFiles/Makefile2:68：CMakeFiles/image_classification_demo.dir/all] 错误 2
make: *** [Makefile:84：all] 错误 2
./run.sh:行19: ./image_classification_demo: 没有那个文件或目录

这个怎么办，树莓派环境

树莓派demo里的目标识别，之前用PaddleCV/SSD的源码运行出的模型替换demo的模型会识别错乱

所以想要一份这个demo的PaddlePaddle源码，然后去学习一下如何从PaddlePaddle到PaddPaddlelite完整的部署一套自己的项目

pi@raspberrypi:~/cmake-3.10.3/Paddle-Lite-Demo/PaddleLite-armlinux-demo/object_detection_demo $ sh run.sh
-- The C compiler identification is unknown
-- The CXX compiler identification is unknown
CMake Error at CMakeLists.txt:16 (project):
The CMAKE_C_COMPILER:

aarch64-linux-gnu-gcc

is not a full path and was not found in the PATH.

Tell CMake where to find the compiler by setting either the environment
variable "CC" or the CMake cache entry CMAKE_C_COMPILER to the full path to
the compiler, or to the compiler name if it is in the PATH.

CMake Error at CMakeLists.txt:16 (project):
The CMAKE_CXX_COMPILER:

aarch64-linux-gnu-g++

is not a full path and was not found in the PATH.

Tell CMake where to find the compiler by setting either the environment
variable "CXX" or the CMake cache entry CMAKE_CXX_COMPILER to the full path
to the compiler, or to the compiler name if it is in the PATH.

-- Configuring incomplete, errors occurred!
See also "/home/pi/cmake-3.10.3/Paddle-Lite-Demo/PaddleLite-armlinux-demo/object_detection_demo/build/CMakeFiles/CMakeOutput.log".
See also "/home/pi/cmake-3.10.3/Paddle-Lite-Demo/PaddleLite-armlinux-demo/object_detection_demo/build/CMakeFiles/CMakeError.log".
make: *** 没有指明目标并且找不到 makefile。 停止。
run.sh: 19: run.sh: ./object_detection_demo: not found

pi@raspberrypi:~/cmake-3.10.3/Paddle-Lite-Demo/PaddleLite-armlinux-demo/image_classification_demo $ sh run.sh
-- The C compiler identification is unknown
-- The CXX compiler identification is unknown
CMake Error at CMakeLists.txt:16 (project):
The CMAKE_C_COMPILER:

aarch64-linux-gnu-gcc

is not a full path and was not found in the PATH.

Tell CMake where to find the compiler by setting either the environment
variable "CC" or the CMake cache entry CMAKE_C_COMPILER to the full path to
the compiler, or to the compiler name if it is in the PATH.

CMake Error at CMakeLists.txt:16 (project):
The CMAKE_CXX_COMPILER:

aarch64-linux-gnu-g++

is not a full path and was not found in the PATH.

Tell CMake where to find the compiler by setting either the environment
variable "CXX" or the CMake cache entry CMAKE_CXX_COMPILER to the full path
to the compiler, or to the compiler name if it is in the PATH.

-- Configuring incomplete, errors occurred!
See also "/home/pi/cmake-3.10.3/Paddle-Lite-Demo/PaddleLite-armlinux-demo/image_classification_demo/build/CMakeFiles/CMakeOutput.log".
See also "/home/pi/cmake-3.10.3/Paddle-Lite-Demo/PaddleLite-armlinux-demo/image_classification_demo/build/CMakeFiles/CMakeError.log".
make: *** 没有指明目标并且找不到 makefile。 停止。

环境：树莓派4 （已源码编译Paddle-Lite）

运行：
$ cd Paddle-Lite-Demo/PaddleLite-armlinux-demo/object_detection_demo
$ ./run.sh armv7hf

有如下报错：

[ 50%] Building CXX object CMakeFiles/image_classification_demo.dir/image_classification_demo.cc.o
/home/pi/Documents/paddle/Paddle-Lite-Demo/PaddleLite-armlinux-demo/image_classification_demo/image_classification_demo.cc:15:10: fatal error: paddle_api.h: 没有那个文件或目录
#include "paddle_api.h"
^~~~~~~~~~~~~~
compilation terminated.
make[2]: *** [CMakeFiles/image_classification_demo.dir/build.make:63：CMakeFiles/image_classification_demo.dir/image_classification_demo.cc.o] 错误 1
make[1]: *** [CMakeFiles/Makefile2:68：CMakeFiles/image_classification_demo.dir/all] 错误 2
make: *** [Makefile:84：all] 错误 2
./run.sh:行19: ./image_classification_demo: 没有那个文件或目录

在android stduio上试图运行ssd_detection_demo， 却出现了下面的错误
A problem occurred configuring project ':app'.

我在服务器上用ssdlite_mobilenetv3_large训练了自己的数据集，然后在服务器上的评估，预测都是正常的。之后利用opt（2.3.0）工具将所得__model__和__params__文件转换成nb文件（全程未报错）。

之后想利用官方的pplite_demo中的目标检测部分来测试。更改了object_detection_demo.cc中的INPUT_SHAPE = {1, 3, 320, 320}; INPUT_MEAN= {0.485, 0.456, 0.406};INPUT_STD={0.229, 0.224, 0.225}。
然后执行run.sh，就出现Segmentation fault (core dumped)这个错误。
我看了之前的issue，里面有用yolov3部署的时候出现同样问题，说是yolov3要有两个输入。是否mobiletv3+ssdlite也是同样的问题？
又或者是我还需要对object_detection_demo.cc中调用的api进行更改？

手机是华为荣耀6
配置信息
型号H60-L02
RAM 3GB
android版本4.4.2 API19
打开APP正常，点击 image classification按钮时产生奔溃现象

logcat信息：
08-28 21:28:01.205 29289-29289/com.baidu.paddle.lite.demo I/dalvikvm: Could not find method android.widget.Button.getAutoSizeMaxTextSize, referenced from method android.support.v7.widget.AppCompatButton.getAutoSizeMaxTextSize
08-28 21:28:01.205 29289-29289/com.baidu.paddle.lite.demo W/dalvikvm: VFY: unable to resolve virtual method 14718: Landroid/widget/Button;.getAutoSizeMaxTextSize ()I
08-28 21:28:01.205 29289-29289/com.baidu.paddle.lite.demo I/dalvikvm: Could not find method android.widget.Button.getAutoSizeMinTextSize, referenced from method android.support.v7.widget.AppCompatButton.getAutoSizeMinTextSize
08-28 21:28:01.205 29289-29289/com.baidu.paddle.lite.demo W/dalvikvm: VFY: unable to resolve virtual method 14719: Landroid/widget/Button;.getAutoSizeMinTextSize ()I
08-28 21:28:01.210 29289-29289/com.baidu.paddle.lite.demo I/dalvikvm: Could not find method android.widget.Button.getAutoSizeStepGranularity, referenced from method android.support.v7.widget.AppCompatButton.getAutoSizeStepGranularity
08-28 21:28:01.210 29289-29289/com.baidu.paddle.lite.demo W/dalvikvm: VFY: unable to resolve virtual method 14720: Landroid/widget/Button;.getAutoSizeStepGranularity ()I
08-28 21:28:01.210 29289-29289/com.baidu.paddle.lite.demo I/dalvikvm: Could not find method android.widget.Button.getAutoSizeTextAvailableSizes, referenced from method android.support.v7.widget.AppCompatButton.getAutoSizeTextAvailableSizes
08-28 21:28:01.210 29289-29289/com.baidu.paddle.lite.demo W/dalvikvm: VFY: unable to resolve virtual method 14721: Landroid/widget/Button;.getAutoSizeTextAvailableSizes ()[I
08-28 21:28:01.210 29289-29289/com.baidu.paddle.lite.demo I/dalvikvm: Could not find method android.widget.Button.getAutoSizeTextType, referenced from method android.support.v7.widget.AppCompatButton.getAutoSizeTextType
08-28 21:28:01.210 29289-29289/com.baidu.paddle.lite.demo W/dalvikvm: VFY: unable to resolve virtual method 14722: Landroid/widget/Button;.getAutoSizeTextType ()I
08-28 21:28:01.210 29289-29289/com.baidu.paddle.lite.demo I/dalvikvm: Could not find method android.widget.Button.setAutoSizeTextTypeUniformWithConfiguration, referenced from method android.support.v7.widget.AppCompatButton.setAutoSizeTextTypeUniformWithConfiguration
08-28 21:28:01.210 29289-29289/com.baidu.paddle.lite.demo W/dalvikvm: VFY: unable to resolve virtual method 14728: Landroid/widget/Button;.setAutoSizeTextTypeUniformWithConfiguration (IIII)V
08-28 21:28:01.210 29289-29289/com.baidu.paddle.lite.demo I/dalvikvm: Could not find method android.widget.Button.setAutoSizeTextTypeUniformWithPresetSizes, referenced from method android.support.v7.widget.AppCompatButton.setAutoSizeTextTypeUniformWithPresetSizes
08-28 21:28:01.210 29289-29289/com.baidu.paddle.lite.demo W/dalvikvm: VFY: unable to resolve virtual method 14729: Landroid/widget/Button;.setAutoSizeTextTypeUniformWithPresetSizes ([II)V
08-28 21:28:01.210 29289-29289/com.baidu.paddle.lite.demo I/dalvikvm: Could not find method android.widget.Button.setAutoSizeTextTypeWithDefaults, referenced from method android.support.v7.widget.AppCompatButton.setAutoSizeTextTypeWithDefaults
08-28 21:28:01.210 29289-29289/com.baidu.paddle.lite.demo W/dalvikvm: VFY: unable to resolve virtual method 14730: Landroid/widget/Button;.setAutoSizeTextTypeWithDefaults (I)V
08-28 21:28:01.220 29289-29289/com.baidu.paddle.lite.demo I/dalvikvm: Could not find method android.graphics.Color.getModel, referenced from method android.support.v4.graphics.ColorUtils.compositeColors
08-28 21:28:01.220 29289-29289/com.baidu.paddle.lite.demo W/dalvikvm: VFY: unable to resolve virtual method 1026: Landroid/graphics/Color;.getModel ()Landroid/graphics/ColorSpace$Model;
08-28 21:28:01.255 29289-29289/com.baidu.paddle.lite.demo I/dalvikvm: Could not find method android.graphics.drawable.Drawable.applyTheme, referenced from method android.support.v4.graphics.drawable.DrawableCompat.applyTheme
08-28 21:28:01.255 29289-29289/com.baidu.paddle.lite.demo W/dalvikvm: VFY: unable to resolve virtual method 1199: Landroid/graphics/drawable/Drawable;.applyTheme (Landroid/content/res/Resources$Theme;)V
08-28 21:28:01.255 29289-29289/com.baidu.paddle.lite.demo I/dalvikvm: Could not find method android.graphics.drawable.Drawable.canApplyTheme, referenced from method android.support.v4.graphics.drawable.DrawableCompat.canApplyTheme
08-28 21:28:01.255 29289-29289/com.baidu.paddle.lite.demo W/dalvikvm: VFY: unable to resolve virtual method 1200: Landroid/graphics/drawable/Drawable;.canApplyTheme ()Z
08-28 21:28:01.255 29289-29289/com.baidu.paddle.lite.demo I/dalvikvm: Could not find method android.graphics.drawable.Drawable.getColorFilter, referenced from method android.support.v4.graphics.drawable.DrawableCompat.getColorFilter
08-28 21:28:01.255 29289-29289/com.baidu.paddle.lite.demo W/dalvikvm: VFY: unable to resolve virtual method 1211: Landroid/graphics/drawable/Drawable;.getColorFilter ()Landroid/graphics/ColorFilter;
08-28 21:28:01.255 29289-29289/com.baidu.paddle.lite.demo I/dalvikvm: Could not find method android.graphics.drawable.Drawable.inflate, referenced from method android.support.v4.graphics.drawable.DrawableCompat.inflate
08-28 21:28:01.255 29289-29289/com.baidu.paddle.lite.demo W/dalvikvm: VFY: unable to resolve virtual method 1228: Landroid/graphics/drawable/Drawable;.inflate (Landroid/content/res/Resources;Lorg/xmlpull/v1/XmlPullParser;Landroid/util/AttributeSet;Landroid/content/res/Resources$Theme;)V
08-28 21:28:01.255 29289-29289/com.baidu.paddle.lite.demo I/dalvikvm: Could not find method android.graphics.drawable.Drawable.setHotspot, referenced from method android.support.v4.graphics.drawable.DrawableCompat.setHotspot
08-28 21:28:01.255 29289-29289/com.baidu.paddle.lite.demo W/dalvikvm: VFY: unable to resolve virtual method 1249: Landroid/graphics/drawable/Drawable;.setHotspot (FF)V
08-28 21:28:01.255 29289-29289/com.baidu.paddle.lite.demo I/dalvikvm: Could not find method android.graphics.drawable.Drawable.setHotspotBounds, referenced from method android.support.v4.graphics.drawable.DrawableCompat.setHotspotBounds
08-28 21:28:01.255 29289-29289/com.baidu.paddle.lite.demo W/dalvikvm: VFY: unable to resolve virtual method 1250: Landroid/graphics/drawable/Drawable;.setHotspotBounds (IIII)V
08-28 21:28:01.255 29289-29289/com.baidu.paddle.lite.demo I/dalvikvm: Could not find method android.graphics.drawable.Drawable.setLayoutDirection, referenced from method android.support.v4.graphics.drawable.DrawableCompat.setLayoutDirection
08-28 21:28:01.255 29289-29289/com.baidu.paddle.lite.demo W/dalvikvm: VFY: unable to resolve virtual method 1251: Landroid/graphics/drawable/Drawable;.setLayoutDirection (I)Z
08-28 21:28:01.255 29289-29289/com.baidu.paddle.lite.demo I/dalvikvm: Could not find method android.graphics.drawable.Drawable.setTint, referenced from method android.support.v4.graphics.drawable.DrawableCompat.setTint
08-28 21:28:01.255 29289-29289/com.baidu.paddle.lite.demo W/dalvikvm: VFY: unable to resolve virtual method 1254: Landroid/graphics/drawable/Drawable;.setTint (I)V
08-28 21:28:01.255 29289-29289/com.baidu.paddle.lite.demo I/dalvikvm: Could not find method android.graphics.drawable.Drawable.setTintList, referenced from method android.support.v4.graphics.drawable.DrawableCompat.setTintList
08-28 21:28:01.255 29289-29289/com.baidu.paddle.lite.demo W/dalvikvm: VFY: unable to resolve virtual method 1255: Landroid/graphics/drawable/Drawable;.setTintList (Landroid/content/res/ColorStateList;)V
08-28 21:28:01.255 29289-29289/com.baidu.paddle.lite.demo I/dalvikvm: Could not find method android.graphics.drawable.Drawable.setTintMode, referenced from method android.support.v4.graphics.drawable.DrawableCompat.setTintMode
08-28 21:28:01.255 29289-29289/com.baidu.paddle.lite.demo W/dalvikvm: VFY: unable to resolve virtual method 1256: Landroid/graphics/drawable/Drawable;.setTintMode (Landroid/graphics/PorterDuff$Mode;)V
08-28 21:28:01.255 29289-29289/com.baidu.paddle.lite.demo I/dalvikvm: Could not find method android.widget.CompoundButton.getButtonDrawable, referenced from method android.support.v4.widget.CompoundButtonCompat.getButtonDrawable
08-28 21:28:01.255 29289-29289/com.baidu.paddle.lite.demo W/dalvikvm: VFY: unable to resolve virtual method 14755: Landroid/widget/CompoundButton;.getButtonDrawable ()Landroid/graphics/drawable/Drawable;
08-28 21:28:01.260 29289-29289/com.baidu.paddle.lite.demo I/dalvikvm: Could not find method android.widget.CompoundButton.getButtonTintList, referenced from method android.support.v4.widget.CompoundButtonCompat.getButtonTintList
08-28 21:28:01.260 29289-29289/com.baidu.paddle.lite.demo W/dalvikvm: VFY: unable to resolve virtual method 14756: Landroid/widget/CompoundButton;.getButtonTintList ()Landroid/content/res/ColorStateList;
08-28 21:28:01.260 29289-29289/com.baidu.paddle.lite.demo I/dalvikvm: Could not find method android.widget.CompoundButton.getButtonTintMode, referenced from method android.support.v4.widget.CompoundButtonCompat.getButtonTintMode
08-28 21:28:01.260 29289-29289/com.baidu.paddle.lite.demo W/dalvikvm: VFY: unable to resolve virtual method 14757: Landroid/widget/CompoundButton;.getButtonTintMode ()Landroid/graphics/PorterDuff$Mode;
08-28 21:28:01.260 29289-29289/com.baidu.paddle.lite.demo I/dalvikvm: Could not find method android.widget.CompoundButton.setButtonTintList, referenced from method android.support.v4.widget.CompoundButtonCompat.setButtonTintList
08-28 21:28:01.260 29289-29289/com.baidu.paddle.lite.demo W/dalvikvm: VFY: unable to resolve virtual method 14773: Landroid/widget/CompoundButton;.setButtonTintList (Landroid/content/res/ColorStateList;)V
08-28 21:28:01.260 29289-29289/com.baidu.paddle.lite.demo I/dalvikvm: Could not find method android.widget.CompoundButton.setButtonTintMode, referenced from method android.support.v4.widget.CompoundButtonCompat.setButtonTintMode
08-28 21:28:01.260 29289-29289/com.baidu.paddle.lite.demo W/dalvikvm: VFY: unable to resolve virtual method 14774: Landroid/widget/CompoundButton;.setButtonTintMode (Landroid/graphics/PorterDuff$Mode;)V
08-28 21:28:02.450 29289-29289/com.baidu.paddle.lite.demo E/dalvikvm: dlopen("/data/app-lib/com.baidu.paddle.lite.demo-1/libpaddle_lite_jni.so") failed: dlopen failed: could not load library "libhiai.so" needed by "libpaddle_lite_jni.so"; caused by cannot locate symbol "__register_atfork" referenced by "libhiai.so"...
08-28 21:28:02.450 29289-29289/com.baidu.paddle.lite.demo W/dalvikvm: Exception Ljava/lang/UnsatisfiedLinkError; thrown while initializing Lcom/baidu/paddle/lite/PaddlePredictor;
08-28 21:28:02.450 29289-29289/com.baidu.paddle.lite.demo W/dalvikvm: threadid=1: thread exiting with uncaught exception (group=0x4159bc50)

mask_detection_demo

在该Demo中，我一直找不到传入Bitmap的入口，和获取预测结果的方法，请问如何得到这个。

使用PaddleX\deploy\lite\android\demo的例子
这个例子直接可以跑起来，但是使用飞桨GUI的视盘分割例子
模型选择 DeepLabV3+ MobileNetV2_x2.0 IMAGENET 512*512

1、生成模型后，导出到移动端
替换model.nb,model.yml,和图片到demo中执行，窗口闪一下就退出
2、如果使用opt模式导出model.nb,model.yml，窗口能打开，但是点击预测按钮退出
使用--print_model_ops=true检测
./opt --model_file=./inference_model/model --param_file=./inference_model/params --valid_targets=arm --optimize_out_type=naive_buffer --optimize_out=tttt --print_model_ops=true
OPs in the input model include:
OP_name Host X86 CUDA ARM OpenCL FPGA NPU XPU RKNPU APU Any Unk
arg_max Y
batch_norm Y Y
bilinear_interp Y Y Y
concat Y Y Y Y
conv2d Y Y Y Y Y
depthwise_conv2d Y Y Y Y Y
dropout Y Y Y Y Y
elementwise_add Y Y Y Y Y
feed Y Y Y
fetch Y Y
reduce_mean Y
relu Y Y Y Y
relu6 Y Y
scale Y Y Y Y Y
shape Y Y
sigmoid Y Y
slice Y Y Y
transpose2 Y Y Y Y
unsqueeze2 Y
Paddle-Lite supports this model!

模型转换，用cxx运行正常，安卓demo预测过程报错

编译结果：

运行结果：

paddle-lite版本 ：HEAD detached at origin/release/v2.0.0-beta1
编译指令为：

./lite/tools/build.sh        \
>     --arm_os=android         \
>     --arm_abi=armv8          \
>     --arm_lang=gcc           \
>     --android_stl=c++_static \
>     tiny_publish

手机型号：小米8 MIUI10.3.6
现象：apk编译安装完成后，打开直接闪退，

报错的log显示如下，这可能是什么原因引起的？
2020-05-22 10:24:45.477 2706-2734/? E/cv::error(): OpenCV(4.2.0) Error: Assertion failed (!ssize.empty()) in resize, file /build/master_pack-android/opencv/modules/imgproc/src/resize.cpp, line 4045
2020-05-22 10:24:45.477 2706-2734/? A/libc: /usr/local/google/buildbot/src/android/ndk-release-r21/external/libcxx/../../external/libcxxabi/src/abort_message.cpp:72: abort_message: assertion "terminating with uncaught exception of type cv::Exception: OpenCV(4.2.0) /build/master_pack-android/opencv/modules/imgproc/src/resize.cpp:4045: error: (-215:Assertion failed) !ssize.empty() in function 'resize'
" failed
2020-05-22 10:24:45.477 2706-2734/? A/libc: Fatal signal 6 (SIGABRT), code -6 in tid 2734 (GLThread 175)
2020-05-22 10:24:45.543 2775-2775/? A/DEBUG: *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** ***
2020-05-22 10:24:45.543 2775-2775/? A/DEBUG: Build fingerprint: 'Android/rk3288/rk3288:7.1.2/NHG47K/lb09161645:userdebug/test-keys'
2020-05-22 10:24:45.543 2775-2775/? A/DEBUG: Revision: '0'
2020-05-22 10:24:45.543 2775-2775/? A/DEBUG: ABI: 'arm'
2020-05-22 10:24:45.543 2775-2775/? A/DEBUG: pid: 2706, tid: 2734, name: GLThread 175 >>> com.baidu.paddle.lite.demo.mask_detection <<<
2020-05-22 10:24:45.544 2775-2775/? A/DEBUG: signal 6 (SIGABRT), code -6 (SI_TKILL), fault addr --------
2020-05-22 10:24:45.544 2775-2775/? A/DEBUG: Abort message: '/usr/local/google/buildbot/src/android/ndk-release-r21/external/libcxx/../../external/libcxxabi/src/abort_message.cpp:72: abort_message: assertion "terminating with uncaught exception of type cv::Exception: OpenCV(4.2.0) /build/master_pack-android/opencv/modules/imgproc/src/resize.cpp:4045: error: (-215:Assertion failed) !ssize.empty() in function 'resize'
" failed'
2020-05-22 10:24:45.544 2775-2775/? A/DEBUG: r0 00000000 r1 00000aae r2 00000006 r3 00000008
2020-05-22 10:24:45.544 2775-2775/? A/DEBUG: r4 9007f978 r5 00000006 r6 9007f920 r7 0000010c
2020-05-22 10:24:45.544 2775-2775/? A/DEBUG: r8 90476d55 r9 ffffff29 sl 00000000 fp 9007e9ec
2020-05-22 10:24:45.544 2775-2775/? A/DEBUG: ip 00000000 sp 9007ce18 lr abb93857 pc abb960c0 cpsr 600f0010
2020-05-22 10:24:45.552 2775-2775/? A/DEBUG: backtrace:
2020-05-22 10:24:45.552 2775-2775/? A/DEBUG: #00 pc 0004a0c0 /system/lib/libc.so (tgkill+12)
2020-05-22 10:24:45.552 2775-2775/? A/DEBUG: #1 pc 00047853 /system/lib/libc.so (pthread_kill+34)
2020-05-22 10:24:45.552 2775-2775/? A/DEBUG: #2 pc 0001d8b5 /system/lib/libc.so (raise+10)
2020-05-22 10:24:45.552 2775-2775/? A/DEBUG: #3 pc 00019401 /system/lib/libc.so (__libc_android_abort+34)
2020-05-22 10:24:45.553 2775-2775/? A/DEBUG: #4 pc 00017048 /system/lib/libc.so (abort+4)
2020-05-22 10:24:45.553 2775-2775/? A/DEBUG: #5 pc 0001b8b3 /system/lib/libc.so (__libc_fatal+22)
2020-05-22 10:24:45.553 2775-2775/? A/DEBUG: #6 pc 000195fb /system/lib/libc.so (__assert2+18)
2020-05-22 10:24:45.553 2775-2775/? A/DEBUG: #7 pc 000645cb /data/app/com.baidu.paddle.lite.demo.mask_detection-1/lib/arm/libc++_shared.so
2020-05-22 10:24:45.553 2775-2775/? A/DEBUG: #8 pc 00064745 /data/app/com.baidu.paddle.lite.demo.mask_detection-1/lib/arm/libc++_shared.so
2020-05-22 10:24:45.553 2775-2775/? A/DEBUG: #9 pc 0006bdf9 /data/app/com.baidu.paddle.lite.demo.mask_detection-1/lib/arm/libc++_shared.so
2020-05-22 10:24:45.553 2775-2775/? A/DEBUG: #10 pc 0006b6a7 /data/app/com.baidu.paddle.lite.demo.mask_detection-1/lib/arm/libc++_shared.so
2020-05-22 10:24:45.553 2775-2775/? A/DEBUG: #11 pc 0006b66f /data/app/com.baidu.paddle.lite.demo.mask_detection-1/lib/arm/libc++_shared.so (__cxa_throw+74)
2020-05-22 10:24:45.553 2775-2775/? A/DEBUG: #12 pc 000878a1 /data/app/com.baidu.paddle.lite.demo.mask_detection-1/lib/arm/libNative.so (_ZN2cv5errorERKNS_9ExceptionE+264)
2020-05-22 10:24:45.553 2775-2775/? A/DEBUG: #13 pc 0008745f /data/app/com.baidu.paddle.lite.demo.mask_detection-1/lib/arm/libNative.so (_ZN2cv5errorEiRKNSt6__ndk112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEPKcSA_i+86)
2020-05-22 10:24:45.553 2775-2775/? A/DEBUG: #14 pc 00141daf /data/app/com.baidu.paddle.lite.demo.mask_detection-1/lib/arm/libNative.so (_ZN2cv6resizeERKNS_11_InputArrayERKNS_12_OutputArrayENS_5Size_IiEEddi+634)
2020-05-22 10:24:45.553 2775-2775/? A/DEBUG: #15 pc 0004b223 /data/app/com.baidu.paddle.lite.demo.mask_detection-1/lib/arm/libNative.so

运行lite安卓yolo detection demo
成功在手机端安装上app，但是闪退，以下是安卓studio build的信息，未报错。
感觉应该是这里的问题，但是才接触astudio，不知道什么意思
ABIs [arm64-v8a,armeabi-v7a,armeabi] set by 'android.injected.build.abi' gradle flag contained 'ARMEABI, ARMEABI_V7A' not targeted by this project.
ABIs [arm64-v8a,armeabi-v7a,armeabi] set by 'android.injected.build.abi' gradle flag contained 'ARMEABI, ARMEABI_V7A' not targeted by this project.

build output 如下：
Executing tasks: [:app:assembleDebug] in project /home/hjs/Paddle-Lite-Demo/PaddleLite-android-demo/yolo_detection_demo

Configure project :app
ABIs [arm64-v8a,armeabi-v7a,armeabi] set by 'android.injected.build.abi' gradle flag contained 'ARMEABI, ARMEABI_V7A' not targeted by this project.
ABIs [arm64-v8a,armeabi-v7a,armeabi] set by 'android.injected.build.abi' gradle flag contained 'ARMEABI, ARMEABI_V7A' not targeted by this project.

Task :app:downloadAndExtractArchives
Downloading and extracting archives including libs and models

Task :app:preBuild
Task :app:preDebugBuild UP-TO-DATE
Task :app:compileDebugAidl NO-SOURCE
Task :app:compileDebugRenderscript NO-SOURCE
Task :app:checkDebugManifest UP-TO-DATE
Task :app:generateDebugBuildConfig UP-TO-DATE
Task :app:prepareLintJar UP-TO-DATE
Task :app:generateDebugSources UP-TO-DATE
Task :app:javaPreCompileDebug UP-TO-DATE
Task :app:mainApkListPersistenceDebug UP-TO-DATE
Task :app:generateDebugResValues UP-TO-DATE
Task :app:generateDebugResources UP-TO-DATE
Task :app:mergeDebugResources UP-TO-DATE
Task :app:createDebugCompatibleScreenManifests UP-TO-DATE
Task :app:processDebugManifest UP-TO-DATE
Task :app:processDebugResources UP-TO-DATE
Task :app:compileDebugJavaWithJavac UP-TO-DATE
Task :app:generateJsonModelDebug UP-TO-DATE

Task :app:externalNativeBuildDebug
Build Native arm64-v8a
ninja: no work to do.

Task :app:compileDebugSources
Task :app:mergeDebugShaders UP-TO-DATE
Task :app:compileDebugShaders UP-TO-DATE
Task :app:generateDebugAssets UP-TO-DATE
Task :app:mergeDebugAssets UP-TO-DATE
Task :app:checkDebugDuplicateClasses UP-TO-DATE
Task :app:mergeExtDexDebug UP-TO-DATE
Task :app:transformClassesWithDexBuilderForDebug UP-TO-DATE
Task :app:mergeDexDebug UP-TO-DATE
Task :app:validateSigningDebug UP-TO-DATE
Task :app:signingConfigWriterDebug UP-TO-DATE
Task :app:mergeDebugJniLibFolders UP-TO-DATE
Task :app:transformNativeLibsWithMergeJniLibsForDebug UP-TO-DATE
Task :app:transformNativeLibsWithStripDebugSymbolForDebug UP-TO-DATE
Task :app:processDebugJavaRes NO-SOURCE
Task :app:transformResourcesWithMergeJavaResForDebug UP-TO-DATE
Task :app:packageDebug UP-TO-DATE
Task :app:assembleDebug

Deprecated Gradle features were used in this build, making it incompatible with Gradle 6.0.
Use '--warning-mode all' to show the individual deprecation warnings.
See https://docs.gradle.org/5.1.1/userguide/command_line_interface.html#sec:command_line_warnings

BUILD SUCCESSFUL in 0s
29 actionable tasks: 2 executed, 27 up-to-date

Build Analyzer results available

你好 我用paddlehub迁移模型mobilenet后再用opt工具转成naive_buffer格式，将转后的best_model.nb和对应的标签文件替换到paddlelite image_classification_demo apk,安装到手机预测结果错误。。 请大神指导一下
我的步骤：
1.cd PaddleHub/demo/image_classification
2../run_classifier.sh --module=mobilenet --checkpoint_dir=./checkpoint --use_gpu=0
3./run_predict.sh --module=mobilenet --checkpoint_dir=./checkpoint --use_gpu=0，预测自下载玫瑰图片为玫瑰，正常。
4../../../Paddle-Lite/opt --model_dir=./checkpoint/best_model
--valid_targets=arm
--optimize_out_type=naive_buffer
--optimize_out=/share/models/best_model

lable:~/.paddlehub/dataset/flower_photos/label_list.txt
5.替换paddlelite image_classification_demo的model.nb和标签文件，生成apk.
6.apk安装在arm8手机，预测自下载玫瑰图片为蒲公英，apk安装在arm7板，预测自下载玫瑰图片为郁金香。 异常。预测结果不对并且同一个apk同一图片在不同机型预测结果不一致！

Android应用口罩识别业务模型集成问题

前言

本仓库中的android demo已成功下载并且在手机上正常运行。
手机型号 - 小米MIX2S
安卓系统 - Lineage os Android 9

问题

目前接到需求，在android工程里集成口罩识别业务。将Paddle-Lite-Demo仓库中的android工程下载后，成功在android studio中编译运行在手机上。而后观察项目结构：

1. 在app module下的build.gradle文件中有两段下载的业务代码，个人分析，是将下载地址中的Paddle-Lite的引擎与业务算法的nb文件下载至cache文件夹中；
2. 参照官网的Java使用文档说明，可以得知，PaddlePredictor起到初始化作用，模仿demo中的工程文件结构进行修改，将app module下的build.gradle的下载地址修改为 https://paddle-inference-dist.bj.bcebos.com/mask_detection.tar.gz ，这是口罩识别C++Demo中列出的口罩识别模型的下载地址，同时将cache文件夹、assets/model文件夹中的内容删除；
3. 以上工作完成后，点击android studio中的同步，项目正常编译完成，点击运行，可正常安装至手机上，但app立马闪退，编译器中无应用层日志输出。逐行调试代码，发现发生闪退处在PaddlePredictor.createPaddlePredictor(config)，此为引擎初始化的操作；

推测

观察原face_detection_demo工程的assets目录，如下图所示：

修改后的mask_detection_demo工程的assets目录，如下图所示：

可以看到区别，口罩的工程里，model文件夹里面有两个目录，Paddle-Lite的开发人员告诉我说，口罩识别模型由两个模型组成，即人脸识别+口罩分类，而仓库中的demo都只有一个独立模型，所以个人推测，直接进行修改，因为目录结构的问题会导致异常产生崩溃。

麻烦开发小哥看一下是不是我推测这样。

Android目标检测时图片大小只有300*300，返回的图片变形，变小，设置shape大小后程序崩溃或者框框不准确，怎么可以返回原图

什么时候出armlinux的python demo？

在Predictor.java中loadModel函数里的

MobileConfig config = new MobileConfig();
config.setModelDir(realPath);

后加上开启多线程预测

config.setThreads(4);

使用opt工具转换了人像分割模型，替换demo中的model.nb之后无法启动，会闪退，机型小米8和一加6都试过
错误信息为
A/libc: Fatal signal 11 (SIGSEGV), code 1 (SEGV_MAPERR), fault addr 0x0 in tid 23429 (Predictor Worke), pid 23399 (an_segmentation)

opt 版本2.3.0和2.6.0都试过
（模型地址：https://paddleseg.bj.bcebos.com/humanseg/models/humanseg_lite_inference.zip）

CMake Error at E:\code\AI\Paddle-Lite-Demo\PaddleLite-android-demo\yolo_detection_demo\app\src\main\cpp\CMakeLists.txt:17 (find_package):
By not providing "FindOpenCV.cmake" in CMAKE_MODULE_PATH this project has
asked CMake to find a package configuration file provided by "OpenCV", but
CMake did not find one.

Could not find a package configuration file provided by "OpenCV" with any
of the following names:

OpenCVConfig.cmake
opencv-config.cmake

Add the installation prefix of "OpenCV" to CMAKE_PREFIX_PATH or set
"OpenCV_DIR" to a directory containing one of the above files. If "OpenCV"
provides a separate development package or SDK, be sure it has been
installed.

demo文件如下（修改了resize和set_float_data）：
from future import absolute_import
from future import division
from future import print_function

import argparse

from paddlelite.lite import *

Command arguments

parser = argparse.ArgumentParser()
parser.add_argument(
"--model_dir", default="", type=str, help="Non-combined Model dir path")

def RunModel(args):
# 1. Set config information
config = MobileConfig()
config.set_model_from_file(args.model_dir)

# 2. Create paddle predictor
predictor = create_paddle_predictor(config)

# 3. Set input data
input_tensor = predictor.get_input(0)
input_tensor.resize([1, 3, 608, 608])
input_tensor.set_float_data([1.] * 3 * 608 * 608)

# 4. Run model
predictor.run()

# 5. Get output data
output_tensor = predictor.get_output(0)
print(output_tensor.shape())
print(output_tensor.float_data()[:10])

if name == 'main':
args = parser.parse_args()
RunModel(args)

调用命令：
python mobilenetv1_light_api.py --model_dir='/home/firefly/Paddle-Lite-Demo-master/PaddleLite-armlinux-demo/object_detection_demo/models/pedestrian_yolov3_darknet/pedestrian_yolov3_darknet_opt.nb'

结果：
[I 7/24 16:32:42.922 ...fly/Paddle-Lite/lite/core/device_info.cc:1064 Setup] ARM multiprocessors name:
[I 7/24 16:32:42.922 ...fly/Paddle-Lite/lite/core/device_info.cc:1065 Setup] ARM multiprocessors number: 6
[I 7/24 16:32:42.922 ...fly/Paddle-Lite/lite/core/device_info.cc:1067 Setup] ARM multiprocessors ID: 0, max freq: 1416, min freq: 1416, cluster ID: 1, CPU ARCH: A53
[I 7/24 16:32:42.922 ...fly/Paddle-Lite/lite/core/device_info.cc:1067 Setup] ARM multiprocessors ID: 1, max freq: 1416, min freq: 1416, cluster ID: 1, CPU ARCH: A53
[I 7/24 16:32:42.922 ...fly/Paddle-Lite/lite/core/device_info.cc:1067 Setup] ARM multiprocessors ID: 2, max freq: 1416, min freq: 1416, cluster ID: 1, CPU ARCH: A53
[I 7/24 16:32:42.922 ...fly/Paddle-Lite/lite/core/device_info.cc:1067 Setup] ARM multiprocessors ID: 3, max freq: 1416, min freq: 1416, cluster ID: 1, CPU ARCH: A53
[I 7/24 16:32:42.922 ...fly/Paddle-Lite/lite/core/device_info.cc:1067 Setup] ARM multiprocessors ID: 4, max freq: 1800, min freq: 1800, cluster ID: 0, CPU ARCH: A72
[I 7/24 16:32:42.922 ...fly/Paddle-Lite/lite/core/device_info.cc:1067 Setup] ARM multiprocessors ID: 5, max freq: 1800, min freq: 1800, cluster ID: 0, CPU ARCH: A72
[I 7/24 16:32:42.922 ...fly/Paddle-Lite/lite/core/device_info.cc:1073 Setup] L1 DataCache size is:
[I 7/24 16:32:42.923 ...fly/Paddle-Lite/lite/core/device_info.cc:1075 Setup] 32 KB
[I 7/24 16:32:42.923 ...fly/Paddle-Lite/lite/core/device_info.cc:1075 Setup] 32 KB
[I 7/24 16:32:42.923 ...fly/Paddle-Lite/lite/core/device_info.cc:1075 Setup] 32 KB
[I 7/24 16:32:42.923 ...fly/Paddle-Lite/lite/core/device_info.cc:1075 Setup] 32 KB
[I 7/24 16:32:42.923 ...fly/Paddle-Lite/lite/core/device_info.cc:1075 Setup] 32 KB
[I 7/24 16:32:42.923 ...fly/Paddle-Lite/lite/core/device_info.cc:1075 Setup] 32 KB
[I 7/24 16:32:42.923 ...fly/Paddle-Lite/lite/core/device_info.cc:1077 Setup] L2 Cache size is:
[I 7/24 16:32:42.923 ...fly/Paddle-Lite/lite/core/device_info.cc:1079 Setup] 512 KB
[I 7/24 16:32:42.923 ...fly/Paddle-Lite/lite/core/device_info.cc:1079 Setup] 512 KB
[I 7/24 16:32:42.923 ...fly/Paddle-Lite/lite/core/device_info.cc:1079 Setup] 512 KB
[I 7/24 16:32:42.923 ...fly/Paddle-Lite/lite/core/device_info.cc:1079 Setup] 512 KB
[I 7/24 16:32:42.923 ...fly/Paddle-Lite/lite/core/device_info.cc:1079 Setup] 512 KB
[I 7/24 16:32:42.923 ...fly/Paddle-Lite/lite/core/device_info.cc:1079 Setup] 512 KB
[I 7/24 16:32:42.923 ...fly/Paddle-Lite/lite/core/device_info.cc:1081 Setup] L3 Cache size is:
[I 7/24 16:32:42.923 ...fly/Paddle-Lite/lite/core/device_info.cc:1083 Setup] 0 KB
[I 7/24 16:32:42.923 ...fly/Paddle-Lite/lite/core/device_info.cc:1083 Setup] 0 KB
[I 7/24 16:32:42.923 ...fly/Paddle-Lite/lite/core/device_info.cc:1083 Setup] 0 KB
[I 7/24 16:32:42.923 ...fly/Paddle-Lite/lite/core/device_info.cc:1083 Setup] 0 KB
[I 7/24 16:32:42.923 ...fly/Paddle-Lite/lite/core/device_info.cc:1083 Setup] 0 KB
[I 7/24 16:32:42.923 ...fly/Paddle-Lite/lite/core/device_info.cc:1083 Setup] 0 KB
[I 7/24 16:32:42.923 ...fly/Paddle-Lite/lite/core/device_info.cc:1085 Setup] Total memory: 3937208KB
[W 7/24 16:32:42.924 ...e-Lite/lite/model_parser/model_parser.cc:811 LoadModelNaiveFromFile] warning: the version of opt that transformed this model is not consistent with current Paddle-Lite version.
version of opt:v2.6.1
version of current Paddle-Lite:00bd7f44
Segmentation fault (core dumped)

是从inference_model和model_with_code中的文件转换得到的吗?

Paddle-Lite-Demo-master\PaddleLite-android-demo\mask_detection_demo在android5.1.1系统运行失败，并报错：
08-17 15:46:00.825 2166-2166/? E/art: dlopen("/data/app/com.baidu.paddle.lite.demo.mask_detection-1/lib/arm/libNative.so", RTLD_LAZY) failed: dlopen failed: cannot locate symbol "__register_atfork" referenced by "libhiai_ir.so"...
08-17 15:46:00.826 2166-2166/? D/AndroidRuntime: Shutting down VM
08-17 15:46:00.827 2166-2166/? E/AndroidRuntime: FATAL EXCEPTION: main
Process: com.baidu.paddle.lite.demo.mask_detection, PID: 2166
java.lang.UnsatisfiedLinkError: dlopen failed: cannot locate symbol "__register_atfork" referenced by "libhiai_ir.so"...
at java.lang.Runtime.loadLibrary(Runtime.java:371)
at java.lang.System.loadLibrary(System.java:988)
at com.baidu.paddle.lite.demo.mask_detection.Native.(Native.java:7)
at com.baidu.paddle.lite.demo.mask_detection.MainActivity.(MainActivity.java:40)
at java.lang.reflect.Constructor.newInstance(Native Method)
at java.lang.Class.newInstance(Class.java:1606)
at android.app.Instrumentation.newActivity(Instrumentation.java:1066)
at android.app.ActivityThread.performLaunchActivity(ActivityThread.java:2242)
at android.app.ActivityThread.handleLaunchActivity(ActivityThread.java:2407)
at android.app.ActivityThread.access$800(ActivityThread.java:151)
at android.app.ActivityThread$H.handleMessage(ActivityThread.java:1319)
at android.os.Handler.dispatchMessage(Handler.java:102)
at android.os.Looper.loop(Looper.java:135)
at android.app.ActivityThread.main(ActivityThread.java:5280)
at java.lang.reflect.Method.invoke(Native Method)
at java.lang.reflect.Method.invoke(Method.java:372)
at com.android.internal.os.ZygoteInit$MethodAndArgsCaller.run(ZygoteInit.java:963)
at com.android.internal.os.ZygoteInit.main(ZygoteInit.java:758)
软硬件环境：
rk3288，android5.1.1，ndk-r21.
但是，完全同样的demo运行在android8.1上没有上述错误。

有的目标检测场景对时间要求稍微宽松，此时FasterRCNN预测效果比YoloV3好很多，目前还没有FasterRCNN的demo示例，如果要部署需要怎么做？

在nano pc t4上用mobilenetv3_yolov3进行目标检测，在预测的时候会出现内存不足的情况，但是后台却显示内存没有用起来。
代码是参照给的安卓demo写的用于arm的代码。
所用的mobilenetv3_yolov3的模型在PC上可以进行检测，之后用opt转换成nb格式，使用自己所写代码就会遇到下面的问题：
[W 12/ 1 10:47:50.325 ...e-Lite/lite/model_parser/model_parser.cc:811 LoadModelNaiveFromFile] warning: the version of opt that transformed this model is not consistent with current Paddle-Lite version. version of opt:v2.6.1 version of current Paddle-Lite:0d53f5af [F 12/ 1 10:47:50.970 ...odel_parser/naive_buffer/naive_buffer.cc:30 Consume] Check failed: bytes <= free_size(): 18874368!<=16687912 No free memory of 18874368, should Require the memory first ./run.sh: line 16: 29699 Aborted (core dumped)

命令：./run.sh armv8
错误：
-- The C compiler identification is GNU 5.4.0
-- The CXX compiler identification is GNU 5.4.0
-- Check for working C compiler: /usr/bin/aarch64-linux-gnu-gcc
-- Check for working C compiler: /usr/bin/aarch64-linux-gnu-gcc -- works
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Detecting C compile features
-- Detecting C compile features - done
-- Check for working CXX compiler: /usr/bin/aarch64-linux-gnu-g++
-- Check for working CXX compiler: /usr/bin/aarch64-linux-gnu-g++ -- works
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- TARGET ARCH ABI: armv8
-- PADDLE LITE DIR: ../Paddle-Lite
-- Found OpenMP_C: -fopenmp (found version "4.0")
-- Found OpenMP_CXX: -fopenmp (found version "4.0")
-- Found OpenMP: TRUE (found version "4.0")
-- Found OpenMP 4.0
-- OpenMP C flags: -fopenmp
-- OpenMP CXX flags: -fopenmp
-- OpenMP OpenMP_CXX_LIB_NAMES: gomp;pthread
-- OpenMP OpenMP_CXX_LIBRARIES: /usr/lib/gcc-cross/aarch64-linux-gnu/5/libgomp.so;/usr/aarch64-linux-gnu/lib/libpthread.so
-- Found OpenCV: /usr/local (found version "3.4.1")
-- OpenCV library status:
-- version: 3.4.1
-- libraries: opencv_calib3d;opencv_core;opencv_dnn;opencv_features2d;opencv_flann;opencv_highgui;opencv_imgcodecs;opencv_imgproc;opencv_ml;opencv_objdetect;opencv_photo;opencv_shape;opencv_stitching;opencv_superres;opencv_video;opencv_videoio;opencv_videostab;opencv_viz
-- include path: /usr/local/include;/usr/local/include/opencv
-- Configuring done
-- Generating done
-- Build files have been written to: /home/user/Desktop/face/Paddle-Lite-Demo/PaddleLite-armlinux-demo/image_classification_demo/build
Scanning dependencies of target image_classification_demo
[ 50%] Building CXX object CMakeFiles/image_classification_demo.dir/image_classification_demo.cc.o
[100%] Linking CXX executable image_classification_demo
/usr/local/lib/libopencv_dnn.so.3.4.1: error adding symbols: File in wrong format
collect2: error: ld returned 1 exit status
CMakeFiles/image_classification_demo.dir/build.make:101: recipe for target 'image_classification_demo' failed
make[2]: *** [image_classification_demo] Error 1
CMakeFiles/Makefile2:72: recipe for target 'CMakeFiles/image_classification_demo.dir/all' failed
make[1]: *** [CMakeFiles/image_classification_demo.dir/all] Error 2
Makefile:83: recipe for target 'all' failed
make: *** [all] Error 2
./run.sh: line 19: ./image_classification_demo: No such file or directory

似乎是与OpenCV有关系，我分别试了OpenCV3.4.1和OpenCV4.3.0，但是都报一样的错误，大神请帮忙看看是什么原因，十分感激。

参考的是给出的https://github.com/PaddlePaddle/Paddle-Lite-Demo/tree/master/PaddleLite-android-demo/object_detection_demo这个目标检测的例子。按照C++ demo中说的方法对模型进行转换，口罩检测的两个模型转换均没有报错，如图
。用pyramidbox_lite_opt.nb模型替换原来目标检测demo中的模型，运行app发现输出结果

麻烦问一下为什么输出为空，是否和预处理有关，谢谢！

2021-03-01 20:53:28.357 17938-17938/? E/AndroidRuntime: FATAL EXCEPTION: main
Process: com.baidu.paddle.lite.demo.yolo_detection, PID: 17938
java.lang.UnsatisfiedLinkError: dlopen failed: library "libhcl.so" not found
at java.lang.Runtime.loadLibrary0(Runtime.java:1016)
at java.lang.System.loadLibrary(System.java:1657)
at com.baidu.paddle.lite.demo.yolo_detection.Native.(Native.java:7)
at com.baidu.paddle.lite.demo.yolo_detection.MainActivity.(MainActivity.java:42)
at java.lang.Class.newInstance(Native Method)
at android.app.Instrumentation.newActivity(Instrumentation.java:1180)
at android.app.ActivityThread.performLaunchActivity(ActivityThread.java:2907)
at android.app.ActivityThread.handleLaunchActivity(ActivityThread.java:3103)
at android.app.ActivityThread.-wrap11(Unknown Source:0)
at android.app.ActivityThread$H.handleMessage(ActivityThread.java:1804)
at android.os.Handler.dispatchMessage(Handler.java:106)
at android.os.Looper.loop(Looper.java:192)
at android.app.ActivityThread.main(ActivityThread.java:6913)
at java.lang.reflect.Method.invoke(Native Method)
at com.android.internal.os.RuntimeInit$MethodAndArgsCaller.run(RuntimeInit.java:549)
at com.android.internal.os.ZygoteInit.main(ZygoteInit.java:828)

Downloading ios-classification_demo/classification_demo/opencv2.framework/Versions/A/opencv2 (145 MB)
Error downloading object: ios-classification_demo/classification_demo/opencv2.framework/Versions/A/opencv2 (cce4b43): Smudge error: Error downloading ios-classification_demo/classification_demo/opencv2.framework/Versions/A/opencv2 (cce4b4336d6136ef4029e9ab9b627212a5520e6393429247fd7c10c491ac6ff7): batch response: This repository is over its data quota. Purchase more data packs to restore access.

Errors logged to /Users/xinnian/workspace/Paddle-Lite-Demo/.git/lfs/logs/20191006T205605.103015.log
Use git lfs logs last to view the log.
error: 外部过滤器 'git-lfs filter-process' 失败
fatal: ios-classification_demo/classification_demo/opencv2.framework/Versions/A/opencv2：smudge 过滤器 lfs 失败

opt优化的模型是2.6.0的，但是PaddleLite-armlinux-demo是2.7版的，模型不能使用，想要PaddleLite-armlinux-demo的2.6.0版本

Solved

PaddleLite-armlinux-demo只有检测和分类，我想在rk3399上进行分割任务，不知道如何下手，我已经完成训练-模型导出-模型转换，现在卡在lite部署，希望能给一定的建议；我见贵方在Android端有分割示例，不知道arm_linux端有类似示例吗

root@huangarm:/home/huang/P-test/Paddle-Lite-Demo/PaddleLite-armlinux-demo/object_detection_demo# ./run.sh
-- The C compiler identification is GNU 9.3.0
-- The CXX compiler identification is GNU 9.3.0
-- Check for working C compiler: /usr/bin/aarch64-linux-gnu-gcc
-- Check for working C compiler: /usr/bin/aarch64-linux-gnu-gcc -- works
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Detecting C compile features
-- Detecting C compile features - done
-- Check for working CXX compiler: /usr/bin/aarch64-linux-gnu-g++
-- Check for working CXX compiler: /usr/bin/aarch64-linux-gnu-g++ -- works
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- TARGET ARCH ABI: armv8
-- PADDLE LITE DIR: ../Paddle-Lite
-- Found OpenMP_C: -fopenmp (found version "4.5")
-- Found OpenMP_CXX: -fopenmp (found version "4.5")
-- Found OpenMP: TRUE (found version "4.5")
-- Found OpenMP 4.5
-- OpenMP C flags: -fopenmp
-- OpenMP CXX flags: -fopenmp
-- OpenMP OpenMP_CXX_LIB_NAMES: gomp;pthread
-- OpenMP OpenMP_CXX_LIBRARIES: /usr/lib/gcc/aarch64-linux-gnu/9/libgomp.so;/usr/lib/aarch64-linux-gnu/libpthread.so
-- Found OpenCV: /usr (found version "4.2.0")
-- OpenCV library status:
-- version: 4.2.0
-- libraries: opencv_calib3d;opencv_core;opencv_dnn;opencv_features2d;opencv_flann;opencv_highgui;opencv_imgcodecs;opencv_imgproc;opencv_ml;opencv_objdetect;opencv_photo;opencv_stitching;opencv_video;opencv_videoio;opencv_aruco;opencv_bgsegm;opencv_bioinspired;opencv_ccalib;opencv_datasets;opencv_dnn_objdetect;opencv_dnn_superres;opencv_dpm;opencv_face;opencv_freetype;opencv_fuzzy;opencv_hdf;opencv_hfs;opencv_img_hash;opencv_line_descriptor;opencv_optflow;opencv_phase_unwrapping;opencv_plot;opencv_quality;opencv_reg;opencv_rgbd;opencv_saliency;opencv_shape;opencv_stereo;opencv_structured_light;opencv_superres;opencv_surface_matching;opencv_text;opencv_tracking;opencv_videostab;opencv_viz;opencv_ximgproc;opencv_xobjdetect;opencv_xphoto
-- include path: /usr/include/opencv4
-- Configuring done
-- Generating done
-- Build files have been written to: /home/huang/P-test/Paddle-Lite-Demo/PaddleLite-armlinux-demo/object_detection_demo/build
Scanning dependencies of target object_detection_demo
[ 50%] Building CXX object CMakeFiles/object_detection_demo.dir/object_detection_demo.cc.o
[100%] Linking CXX executable object_detection_demo
[100%] Built target object_detection_demo
[I 2/19 16:29: 2.778 ...ang/Paddle-Lite/lite/core/device_info.cc:1065 Setup] ARM multiprocessors name: HARDWARE : SUN50IW1P1
HARDWARE : SUN50IW1P1
HARDWARE : SUN50IW1P1
HARDWARE : SUN50IW1P1

[I 2/19 16:29: 2.778 ...ang/Paddle-Lite/lite/core/device_info.cc:1066 Setup] ARM multiprocessors number: 4
[I 2/19 16:29: 2.778 ...ang/Paddle-Lite/lite/core/device_info.cc:1068 Setup] ARM multiprocessors ID: 0, max freq: 1800, min freq: 1800, cluster ID: 0, CPU ARCH: A53
[I 2/19 16:29: 2.778 ...ang/Paddle-Lite/lite/core/device_info.cc:1068 Setup] ARM multiprocessors ID: 1, max freq: 1800, min freq: 1800, cluster ID: 0, CPU ARCH: A53
[I 2/19 16:29: 2.778 ...ang/Paddle-Lite/lite/core/device_info.cc:1068 Setup] ARM multiprocessors ID: 2, max freq: 1800, min freq: 1800, cluster ID: 0, CPU ARCH: A53
[I 2/19 16:29: 2.778 ...ang/Paddle-Lite/lite/core/device_info.cc:1068 Setup] ARM multiprocessors ID: 3, max freq: 1800, min freq: 1800, cluster ID: 0, CPU ARCH: A53
[I 2/19 16:29: 2.778 ...ang/Paddle-Lite/lite/core/device_info.cc:1074 Setup] L1 DataCache size is:
[I 2/19 16:29: 2.778 ...ang/Paddle-Lite/lite/core/device_info.cc:1076 Setup] 32 KB
[I 2/19 16:29: 2.778 ...ang/Paddle-Lite/lite/core/device_info.cc:1076 Setup] 32 KB
[I 2/19 16:29: 2.778 ...ang/Paddle-Lite/lite/core/device_info.cc:1076 Setup] 32 KB
[I 2/19 16:29: 2.778 ...ang/Paddle-Lite/lite/core/device_info.cc:1076 Setup] 32 KB
[I 2/19 16:29: 2.778 ...ang/Paddle-Lite/lite/core/device_info.cc:1078 Setup] L2 Cache size is:
[I 2/19 16:29: 2.778 ...ang/Paddle-Lite/lite/core/device_info.cc:1080 Setup] 512 KB
[I 2/19 16:29: 2.778 ...ang/Paddle-Lite/lite/core/device_info.cc:1080 Setup] 512 KB
[I 2/19 16:29: 2.778 ...ang/Paddle-Lite/lite/core/device_info.cc:1080 Setup] 512 KB
[I 2/19 16:29: 2.778 ...ang/Paddle-Lite/lite/core/device_info.cc:1080 Setup] 512 KB
[I 2/19 16:29: 2.778 ...ang/Paddle-Lite/lite/core/device_info.cc:1082 Setup] L3 Cache size is:
[I 2/19 16:29: 2.778 ...ang/Paddle-Lite/lite/core/device_info.cc:1084 Setup] 0 KB
[I 2/19 16:29: 2.778 ...ang/Paddle-Lite/lite/core/device_info.cc:1084 Setup] 0 KB
[I 2/19 16:29: 2.778 ...ang/Paddle-Lite/lite/core/device_info.cc:1084 Setup] 0 KB
[I 2/19 16:29: 2.778 ...ang/Paddle-Lite/lite/core/device_info.cc:1084 Setup] 0 KB
[I 2/19 16:29: 2.779 ...ang/Paddle-Lite/lite/core/device_info.cc:1086 Setup] Total memory: 1008684KB
[ WARN:0] global ../modules/videoio/src/cap_gstreamer.cpp (1758) handleMessage OpenCV | GStreamer warning: Embedded video playback halted; module v4l2src0 reported: Device '/dev/video0' is not a capture device.
[ WARN:0] global ../modules/videoio/src/cap_gstreamer.cpp (888) open OpenCV | GStreamer warning: unable to start pipeline
[ WARN:0] global ../modules/videoio/src/cap_gstreamer.cpp (480) isPipelinePlaying OpenCV | GStreamer warning: GStreamer: pipeline have not been created
[ WARN:0] global ../modules/videoio/src/cap_v4l.cpp (887) open VIDEOIO(V4L2:/dev/video0): can't open camera by index
root@huangarm:/home/huang/P-test/Paddle-Lite-Demo/PaddleLite-armlinux-demo/object_detection_demo#

环境：Ubuntu16.04
命令：./run.sh armv8
错误信息如下：
-- The C compiler identification is GNU 5.4.0
-- The CXX compiler identification is GNU 5.4.0
-- Check for working C compiler: /usr/bin/aarch64-linux-gnu-gcc
-- Check for working C compiler: /usr/bin/aarch64-linux-gnu-gcc -- works
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Detecting C compile features
-- Detecting C compile features - done
-- Check for working CXX compiler: /usr/bin/aarch64-linux-gnu-g++
-- Check for working CXX compiler: /usr/bin/aarch64-linux-gnu-g++ -- works
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- TARGET ARCH ABI: armv8
-- PADDLE LITE DIR: ../Paddle-Lite
-- Found OpenMP_C: -fopenmp (found version "4.0")
-- Found OpenMP_CXX: -fopenmp (found version "4.0")
-- Found OpenMP: TRUE (found version "4.0")
-- Found OpenMP 4.0
-- OpenMP C flags: -fopenmp
-- OpenMP CXX flags: -fopenmp
-- OpenMP OpenMP_CXX_LIB_NAMES: gomp;pthread
-- OpenMP OpenMP_CXX_LIBRARIES: /usr/lib/gcc-cross/aarch64-linux-gnu/5/libgomp.so;/usr/aarch64-linux-gnu/lib/libpthread.so
-- Found OpenCV: /usr/local (found version "4.3.0")
-- OpenCV library status:
-- version: 4.3.0
-- libraries: opencv_calib3d;opencv_core;opencv_dnn;opencv_features2d;opencv_flann;opencv_gapi;opencv_highgui;opencv_imgcodecs;opencv_imgproc;opencv_ml;opencv_objdetect;opencv_photo;opencv_stitching;opencv_video;opencv_videoio
-- include path: /usr/local/include/opencv4
-- Configuring done
-- Generating done
-- Build files have been written to: /home/user/Desktop/face/Paddle-Lite-Demo/PaddleLite-armlinux-demo/image_classification_demo/build
Scanning dependencies of target image_classification_demo
[ 50%] Building CXX object CMakeFiles/image_classification_demo.dir/image_classification_demo.cc.o
/home/user/Desktop/face/Paddle-Lite-Demo/PaddleLite-armlinux-demo/image_classification_demo/image_classification_demo.cc: In function ‘void preprocess(cv::Mat&, const std::vector&, const std::vector&, int, int, float*)’:
/home/user/Desktop/face/Paddle-Lite-Demo/PaddleLite-armlinux-demo/image_classification_demo/image_classification_demo.cc:70:46: error: ‘CV_BGRA2RGB’ was not declared in this scope
cv::cvtColor(resize_image, resize_image, CV_BGRA2RGB);
^
/home/user/Desktop/face/Paddle-Lite-Demo/PaddleLite-armlinux-demo/image_classification_demo/image_classification_demo.cc: In function ‘cv::Mat process(cv::Mat&, std::vector<std::__cxx11::basic_string >&, std::shared_ptrpaddle::lite_api::PaddlePredictor&)’:
/home/user/Desktop/face/Paddle-Lite-Demo/PaddleLite-armlinux-demo/image_classification_demo/image_classification_demo.cc:211:41: warning: format ‘%d’ expects argument of type ‘int’, but argument 2 has type ‘std::vector::size_type {aka long unsigned int}’ [-Wformat=]
printf("results: %d\n", results.size());
^
/home/user/Desktop/face/Paddle-Lite-Demo/PaddleLite-armlinux-demo/image_classification_demo/image_classification_demo.cc: In function ‘int main(int, char**)’:
/home/user/Desktop/face/Paddle-Lite-Demo/PaddleLite-armlinux-demo/image_classification_demo/image_classification_demo.cc:259:13: error: ‘CV_CAP_PROP_FRAME_WIDTH’ was not declared in this scope
cap.set(CV_CAP_PROP_FRAME_WIDTH, 640);
^
/home/user/Desktop/face/Paddle-Lite-Demo/PaddleLite-armlinux-demo/image_classification_demo/image_classification_demo.cc:260:13: error: ‘CV_CAP_PROP_FRAME_HEIGHT’ was not declared in this scope
cap.set(CV_CAP_PROP_FRAME_HEIGHT, 480);
^
CMakeFiles/image_classification_demo.dir/build.make:62: recipe for target 'CMakeFiles/image_classification_demo.dir/image_classification_demo.cc.o' failed
make[2]: *** [CMakeFiles/image_classification_demo.dir/image_classification_demo.cc.o] Error 1
CMakeFiles/Makefile2:72: recipe for target 'CMakeFiles/image_classification_demo.dir/all' failed
make[1]: *** [CMakeFiles/image_classification_demo.dir/all] Error 2
Makefile:83: recipe for target 'all' failed
make: *** [all] Error 2
./run.sh: line 19: ./image_classification_demo: No such file or directory

请帮忙看看是什么原因导致的错误，谢谢。

如题，我在执行PaddleLite-armlinux-demo/object_detection_demo的./run.sh armv8这一步出现问题，如下：

/home/nvidia/Paddle-Lite-Demo/PaddleLite-armlinux-demo/object_detection_demo/object_detection_demo.cc: In function ‘std::vector<std::__cxx11::basic_string > load_labels(const string&)’:
/home/nvidia/Paddle-Lite-Demo/PaddleLite-armlinux-demo/object_detection_demo/object_detection_demo.cc:52:17: error: aggregate ‘std::ifstream file’ has incomplete type and cannot be defined
std::ifstream file;
^~~~
/home/nvidia/Paddle-Lite-Demo/PaddleLite-armlinux-demo/object_detection_demo/object_detection_demo.cc: In function ‘cv::Mat process(cv::Mat&, std::vector<std::__cxx11::basic_string >&, std::shared_ptrpaddle::lite_api::PaddlePredictor&)’:
/home/nvidia/Paddle-Lite-Demo/PaddleLite-armlinux-demo/object_detection_demo/object_detection_demo.cc:220:41: warning: format ‘%d’ expects argument of type ‘int’, but argument 2 has type ‘std::vector::size_type {aka long unsigned int}’ [-Wformat=]
printf("results: %d\n", results.size());
~~~~~~~~~~~~~~^
CMakeFiles/object_detection_demo.dir/build.make:62: recipe for target 'CMakeFiles/object_detection_demo.dir/object_detection_demo.cc.o' failed
make[2]: *** [CMakeFiles/object_detection_demo.dir/object_detection_demo.cc.o] Error 1
CMakeFiles/Makefile2:67: recipe for target 'CMakeFiles/object_detection_demo.dir/all' failed
make[1]: *** [CMakeFiles/object_detection_demo.dir/all] Error 2
Makefile:83: recipe for target 'all' failed
make: *** [all] Error 2
./run.sh: line 19: ./object_detection_demo: No such file or directory

When I run face_detection_demo, it appears as follows:
Transform output file E:\Androidproject\face\Paddle-Lite-Demo-master\PaddleLite-android-demo\face_detection_demo\app\libs\PaddlePredictor.jar does not exist.
Is this file automatically generated？？

Hi，all
之前我把这个issue 调到了另外的地方，请帮忙定位下问题

设备信息： rk3399 AIO-33399J
OS: android 7
paddle lite 2.7.0

具体请查看下面的链接
PaddlePaddle/Paddle-Lite#5559 (comment)

目的：使用ocr_attention训练出来的预测模型在android手机上运行
操作步骤：
1，将预测模型放在/assets/image_classification/models/test_model_ocr 目录下
2，将测试图片放在/assets/image_classification/images/000.jpg
3，修改strings.xml
` xml

<string name="ICS_MODEL_PATH_DEFAULT">image_classification/labels/test_model_ocr</string>
<string name="ICS_LABEL_PATH_DEFAULT">image_classification/labels/synset_words.txt</string>

<string name="ICS_IMAGE_PATH_DEFAULT">image_classification/images/000.jpg</string>

<string name="ICS_INPUT_SHAPE_DEFAULT">1,3,100,380</string>

`
4，修改了ImgClassifyPredictor的输出读取。

然后debug的时候会报错，错误节点发生在Predictor.java的runModel()方法中
public boolean runModel() { ... paddlePredictors.get(whichDevice).run(); // 报错 ... }

错误截图如下：

你好，我自己训练了blazeface的模型并根据ios_detection_demo为基础进行部署，但部署后竖屏和横屏的预测结果并不一致。这两张图分别是横竖屏状态下的截图，可以看到侦测到的人脸bbox的大小比例并不一致。除了把图片旋转90度来适配横屏之外没有对代码做其他修改。

如题
按键，实时视频，bounding box都能正确显示，但result的detection结果无法正常显示

From the demo Paddle-Lite-Demo , I found that mobilenet_v1_for_cpu and mobilenet_v1_for_npu have different size, so I suppose that they are not the same model.

However, from the description for model_optimize_tool, I did not get the related information, thus I am not sure whether the optimized model generated by mode_optimize_tool is for CPU or NPU.

More concretely, I am interested in the method to generate the model for running on NPU. Could you provide a solution to this need? Thanks in advance.

线程：1
性能模式：LITE_POWER_HIGH
时间：800ms -> 3400ms

荣耀20手机 型号YAL-AL00 android 10 Magic UI 3.0.0
android studio 3.4.1
04/06 12:30:46: Launching app
$ adb install-multiple -r -t /home/zhangshilin/Paddle-Lite-Demo-master/PaddleLite-android-demo/mask_detection_demo/app/build/intermediates/split-apk/debug/slices/slice_8.apk /home/zhangshilin/Paddle-Lite-Demo-master/PaddleLite-android-demo/mask_detection_demo/app/build/intermediates/split-apk/debug/slices/slice_2.apk /home/zhangshilin/Paddle-Lite-Demo-master/PaddleLite-android-demo/mask_detection_demo/app/build/intermediates/split-apk/debug/slices/slice_3.apk /home/zhangshilin/Paddle-Lite-Demo-master/PaddleLite-android-demo/mask_detection_demo/app/build/intermediates/split-apk/debug/slices/slice_1.apk /home/zhangshilin/Paddle-Lite-Demo-master/PaddleLite-android-demo/mask_detection_demo/app/build/intermediates/split-apk/debug/dep/dependencies.apk /home/zhangshilin/Paddle-Lite-Demo-master/PaddleLite-android-demo/mask_detection_demo/app/build/intermediates/split-apk/debug/slices/slice_0.apk /home/zhangshilin/Paddle-Lite-Demo-master/PaddleLite-android-demo/mask_detection_demo/app/build/intermediates/split-apk/debug/slices/slice_5.apk /home/zhangshilin/Paddle-Lite-Demo-master/PaddleLite-android-demo/mask_detection_demo/app/build/intermediates/split-apk/debug/slices/slice_4.apk /home/zhangshilin/Paddle-Lite-Demo-master/PaddleLite-android-demo/mask_detection_demo/app/build/intermediates/resources/instant-run/debug/resources-debug.apk /home/zhangshilin/Paddle-Lite-Demo-master/PaddleLite-android-demo/mask_detection_demo/app/build/intermediates/split-apk/debug/slices/slice_9.apk /home/zhangshilin/Paddle-Lite-Demo-master/PaddleLite-android-demo/mask_detection_demo/app/build/intermediates/split-apk/debug/slices/slice_6.apk /home/zhangshilin/Paddle-Lite-Demo-master/PaddleLite-android-demo/mask_detection_demo/app/build/intermediates/split-apk/debug/slices/slice_7.apk /home/zhangshilin/Paddle-Lite-Demo-master/PaddleLite-android-demo/mask_detection_demo/app/build/intermediates/instant-run-apk/debug/app-debug.apk
Split APKs installed in 10 s 897 ms
$ adb shell am start -n "com.baidu.paddle.lite.demo.mask_detection/com.baidu.paddle.lite.demo.mask_detection.MainActivity" -a android.intent.action.MAIN -c android.intent.category.LAUNCHER
Client not ready yet..Waiting for process to come online
Connected to process 16975 on device huawei-yal_al00-STSDU19529011335
Application terminated.

app run后

手机上mask detection小图标，点击打开，闪退（打不开）？？怎么解？

按照“PaddleLite树莓派从0到1：安全帽检测小车部署（一）修改了run.sh最后一段的内容
#run
LD_LIBRARY_PATH=$LD_LIBRARY_PATH:${PADDLE_LITE_DIR}/libs/${TARGET_ARCH_ABI} ./object_detection_demo ../models/ssd_mobilenet_v1_pascalvoc_for_cpu/ssd_mobilenet_v1_voc.nb ../labels/helmet_label，想要在TX2上调用USB摄像头。
”运行object_detection_demo的run.sh报错如下：

[F 1/20 13:44:22.854 ...engyang/Paddle-Lite/lite/core/program.cc:174 RuntimeProgram] Check failed: block_idx >= 0 && block_idx < block_size: Invalid block index, expected [0,18446744073709551615] but got 0
[F 1/20 13:44:22.854 ...engyang/Paddle-Lite/lite/core/program.cc:174 RuntimeProgram] Check failed: block_idx >= 0 && block_idx < block_size: Invalid block index, expected [0,18446744073709551615] but got 0

[F 1/20 13:44:22.854 ...model_parser/flatbuffers/program_desc.cc:31 GetBlockpaddle::lite::fbs::BlockDescView] Check failed: (idx < static_cast<int32_t>(BlocksSize())): 0!<0 idx >= blocks.size()
[F 1/20 13:44:22.854 ...model_parser/flatbuffers/program_desc.cc:31 GetBlockpaddle::lite::fbs::BlockDescView] Check failed: (idx < static_cast<int32_t>(BlocksSize())): 0!<0 idx >= blocks.size()

./run.sh: line 19: 19493 Segmentation fault (core dumped) LD_LIBRARY_PATH=$LD_LIBRARY_PATH:${PADDLE_LITE_DIR}/libs/${TARGET_ARCH_ABI} ./object_detection_demo ../models/ssd_mobilenet_v1_pascalvoc_for_cpu/ssd_mobilenet_v1_voc.nb ../labels/helmet_label