Introduction
This is a highly separated deployment project based on Deepstream 5.0, including the full range of YOLO and an OCR project. The project will be updated and expanded in the future.
You can get familiar with the plug-in development of deepstream through this project.
Deepstream_Yolo, which contains the engineering method of yolov4 and yolov5 conversion model and the engineering method of deploying deepstream.
Ocr, this is a text recognition deployment project based on YOLOV3 and CRNN, and it is still in the supplementary description.
The above projects are highly separated, but you still have to make sure that your environment configuration, cuda, cudnn, Tensorrt, Deepstream are installed successfully and can be compiled and used, pay attention to your dependent environment variables, dynamic library reference paths and header file references, etc. For specific instructions, please read the README in the respective directory carefully.
这是一个基于Deepstream 5.1的高度分离工程，包含YOLO全系列以及一个OCR工程，后续会补充更新和不断扩充该项目，可以通过该项目去熟悉deepstream的插件化开发.
Deepstream_Yolo,这包含了yolov4 和yolov5的转换模型的工程方法和部署deepstream的工程方法.
Ocr，这是基于YOLOV3和CRNN的文字识别部署工程，还在补充说明中..
以上工程做了高度的分离，但还是要确保你的环境配置,cuda,cudnn,Tensorrt,Deepstream安装成功且可编译使用，注意你的依赖环境变量，动态库的引用路径和头文件的引用等，具体使用说明，请仔细看各自目录下的README
Will be updated and supplemented later
后面会详细更新补充
Reference：
yolov5: https://github.com/ultralytics/yolov5
darknet: https://github.com/AlexeyAB/darknet
tensorrtx:https://github.com/wang-xinyu/tensorrtx