search

HaiderAbasi / OpenCV-Raspberry-Pi-4-Projects-on-the-Edge

Harnessing the power of Raspberry Pi 4 to build cutting-edge computer vision solutions. Whether you're interested in object detection, image classification, or real-time video analysis, this project will give you the tools you need to get started.
MIT License
0 stars 0 forks source link

RPI NCNN library Setup #20

Open noshluk2 opened 1 year ago

noshluk2 commented 1 year ago

Written by @HaiderAbasi
TFLite-> Looks like a no-show. Why? Speed:

1) Video example

ImageVideo

2) Community

ImageStackOverflow

3) YoloV4-tiny on NCNN (Q-Engineer)

Image YoloV4-tiny_NCNN

======================================================================

Why is this a concern?

No speed No tracking as most trackers work on tracking by detection. Meaning they run detection on every frame so slowness overpowered

Why do we need tracking?

We need tracking to Know!

This is the same car that we saw one frame ago This is the same car we counted one frame ago (Car counter) This is the same car that was at point A then and Point B now (Overspeed detection) This is the same car whos identity we have verified . No need to do it again (Number plate recognition) Whats the solution?

Stick to Ncnn Try to give inference speed of each of the top models (YoloX-Nano - NanoDet - SSD) and get viable prediciton resultss Get this working: https://github.com/Qengineering/Traffic-Counter-RPi_64-bit

noshluk2 commented 1 year ago

halting because ncnn is not a a mainstream framework so for people it might be a "NO" thing .