YOLOv9 is better than any convolution or transformer based object detectors

AlexeyAB / darknet

YOLOv4 / Scaled-YOLOv4 / YOLO - Neural Networks for Object Detection (Windows and Linux version of Darknet )

http://pjreddie.com/darknet/

Other

21.75k stars 7.96k forks source link

YOLOv9 is better than any convolution or transformer based object detectors #8887

Open AlexeyAB opened 8 months ago

AlexeyAB commented 8 months ago

YOLOv9 is better than any convolution or transformer based object detectors

Paper: https://arxiv.org/abs/2402.13616
Code: https://github.com/WongKinYiu/yolov9

waytechpro commented 8 months ago

Absolutely! YOLOv9 certainly stands out among object detectors with its efficient architecture and remarkable performance. Its streamlined approach surpasses traditional convolutional and transformer-based models in terms of speed and accuracy. If you're keen on exploring cutting-edge object detection capabilities, YOLOv9 is definitely worth considering. Ready to elevate your detection game? Click here to learn more.

1027663760 commented 8 months ago

Have you considered adapting YOLOV9 for Darknet?

toplinuxsir commented 8 months ago

It is a great idea!

TaihuLight commented 3 months ago

@AlexeyAB @cenit @stephanecharette @toplinuxsir It is better to implement YOLOv9 in this version of darknet rather than the newer one hank-ai/darknet. This is because C programming is more suitable for edge devices than C++.

stephanecharette commented 3 months ago

This is because C programming is more suitable for edge devices than C++.

As someone who uses Darknet/YOLO on edge devices, I strongly disagree.

stephanecharette commented 3 months ago

You may also want to watch this video: https://www.youtube.com/watch?v=2Mq23LFv1aM