The tensorflow_object_detection_tflite from manuelschmitzberger

manuelschmitzberger / tensorflow_object_detection_tflite Goto Github PK

View Code? Open in Web Editor NEW

This is a repo for mobilenet-ssd tflite c++ on x86 and arm64

CMake 0.97% C++ 84.79% Python 4.93% C 3.91% Java 1.12% Shell 1.07% Objective-C 0.19% Objective-C++ 0.47% Jupyter Notebook 1.05% Ruby 0.01% C# 0.53% Swift 0.12% Makefile 0.33% Batchfile 0.03% RobotFramework 0.01% PHP 0.38% Pawn 0.03% Assembly 0.01% Pascal 0.01% Smarty 0.05%

tensorflow_object_detection_tflite's Introduction

System Environment:

System: Ubuntu 18.04

Opencv: opencv 4.1.0

Instructions:

Run demo.cpp on x86 unbuntu, make sure opencv and bazel are installed.
1. Build libtensorflowlite.so, under the tensorflow directory.
```
            bazel build -c opt //tensorflow/lite:libtensorflowlite.so --fat_apk_cpu=arm64-v8a
```
1. Move .so to tensorflow_object_detection_tflite/lib
2. Change find_library(TFLITE_LIBRARY tensorflow-lite "lib") to find_library(TFLITE_LIBRARY tensorflowlite "lib") in CMakeLists.txt.
3. Build cmake
```
        mkdir build
        cd build
        cmake ..
        make -j
        ./demo
```
Run demo.cpp on arm64-v8a ubuntu.
1. Intall opencv on your arm64 motherboard.
2. Build libtensorflow-lite.a, followed by the tensorflow tutorial https://www.tensorflow.org/lite/guide/build_arm64. Careful about the arm version, v7 or v8.
3. Move .a to tensorflow_object_detection_tflite/lib
4. keep find_library(TFLITE_LIBRARY tensorflow-lite "lib") unchanged.
5. Build cmake
```
        mkdir build
        cd build
        cmake ..
        make -j
        ./demo
```
If there is a flatbuffers error, build flatbuffers on your desktop, and use its header files and .a lib file, put and replace them into tensorflow_object_detection_tflite/include and tensorflow_object_detection_tflite/lib, respectively. You can check here to know how to build. google/flatbuffers#5569 (comment)
Result image