Evaluate YoloX_nano ncnn on images and test_videos

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

Evaluate YoloX_nano ncnn on images and test_videos #18

Closed HaiderAbasi closed 1 year ago

HaiderAbasi commented 1 year ago

Notes:

Impressive accuracy for an ncnn model on windows10
speed unknown, preferable to test on pi4
map seems to be the highest of all
vehicle classification is the most stable of all previously tested

Todo:

Speed and accuracy test on pi
develop a car counter using yolox_nano