Introduction

This project implement and deploy many embedded AI framework for edge devices, such as: NCNN, Paddle, MNN, SNPE, and Tensorflow lite, ... Also, develop some tasks: object detection, auto plate recognition, OCR, object tracking.

Requirements

1. paddle-lite 2.9

2. adk r20b

3. Opencv

4. snpe 1.52

std c++14 OpenCV 4.5.0 built from source with c++. sudo apt-get install build-essential cmake pkg-config sudo apt-get install libjpeg-dev libtiff5-dev libavcodec-dev libavformat-dev libswscale-dev libv4l-dev libgstreamer1.0-dev libgstreamer-plugins-base1.0-dev x264 v4l-utils libhdf5-dev libprotobuf-dev protobuf-compiler git libgtk2.0-dev libgtk-3-dev

Build & Run

Some commands are:

mkdir build cd build cmake .. make ./tracker

References

ncnn: https://github.com/Tencent/ncnn snpe: paddle: