Yolo fast+dnn+flask realizes streaming and streaming on mobile terminals and displays them on the web

Reprint please indicate the source ！
Reprint please indicate the source ！
Reprint please indicate the source ！

Project code link ：https://github.com/pengtougu/Push-Streaming.git

It's twoorthree o'clock , Too sleepy , direct github Upper readme copied , The weekend is a little boring , Made a push-pull flow demo, There are four functions ：
① Picture reasoning
② Video reasoning and preservation
③ Camera local reasoning （ Don't save , A little memory consumption ）
④ Mobile （ Raspberry pie , Or other development boards ） Call the camera and reason about the frame , adopt flask Push streaming to LAN , Other devices under the LAN pull the stream and display it on web On the page

The code basically doesn't need to be changed ,down Come down and run , Already in window&mac&linux Tested on three platforms , The code is generic . Push pull flow , Please ensure that you are under the same LAN ！！！

Only now yolo-fastest Of demo, Yes nanodet Interested in , It will be updated continuously in the future ！

Project code link ：https://github.com/pengtougu/Push-Streaming.git

Push-Streaming

Hi, this repository documents the process of pushing streams on some ultra-lightweight nets. The general steps are that opencv calls the board（like Raspberry Pi）'s camera, transmits the detected live video to an ultra-lightweight network like yolo-fastest, nanodet, ghostnet, and then talks about pushing the processed video frames to the web using the flask lightweight framework, which basically guarantees real-time performance.

Requirements

Please install the following packages first

• Linux & MacOS & window
• python>= 3.6.0
• opencv-python>= 4.2.X
• flask>= 1.0.0

inference

• Yolo-Fastest： https://github.com/dog-qiuqiu/Yolo-Fastest
  Models：Yolo-Fastest-1.1-xl

Equipment	Computing backend	System	Framework	Run time
Raspberrypi 3B	4xCortex-A53	Linux(arm64)	dnn	89ms
Intel	Core i5-4210	window10（x64）	dnn	67ms

• Nanodet： https://github.com/RangiLyu/nanodet

  updating. . .

Demo

First of all, I have tested this demo in window, mac and linux environments and it works in all of them.

The students who pull down look at the documents first ：

• Inference images

python yolov3_fastest.py --image dog.jpg

• Inference video

python yolov3_fastest.py --video test.mp4

• Inference webcam

python yolov3_fastest.py --fourcc 0

• Push-Streaming

python app.py
（ Please make sure your raspberry pie has the camera driver installed , And the board is connected with the local machine WiFi）
（ Please make sure your raspberry pie has the camera driver installed , And the board is connected with the local machine WiFi）
（ Please make sure your raspberry pie has the camera driver installed , And the board is connected with the local machine WiFi）

Please note! Be sure to be on the same LAN！

Demo Effects

• Demo images
  

• Demo video
  

• Demo camera
  

• Demo Push-Streaming
  
  Project code link ：https://github.com/pengtougu/Push-Streaming.git

Thanks

• https://github.com/dog-qiuqiu/Yolo-Fastest
• https://github.com/hpc203/Yolo-Fastest-opencv-dnn
• https://github.com/miguelgrinberg/flask-video-streaming
• Thank the boss of this blog for providing ideas ：https://blog.csdn.net/nihate/article/details/108670542

remarks

I haven't blogged for a year , This year, I really met many big guys , Or sigh that there are too many things to learn