Complete YOLO v8 Custom Object Detection Tutorial | Windows & Linux

A complete YOLO v8 custom object detection tutorial with two-classe custom dataset. I cover how to annotate custom dataset in YOLO format, setting up environment for YOLO v8, and train custom object detection model. Although I show a demonstration using YOLO v8 medium version, the method is generic enough to train all variations of YOLO v8.

** Code is available for our Patreon Supporters**
---------------------------------------------
► Time Stamps:
Introduction: (0:00)
Setting up Virtual Environment: (0:35)
Develop Custom Object Detection Dataset: (1:08)
Annotating custom dataset in YOLO format: (3:07)
Setup YOLO v8 repository: (6:14)
Train YOLO v8 custom: (7:54)
Run custom object detector on images: (11:14)
Run custom object detector on videos: (12:25)
Run custom object detector on webcam: (13:56)
Export YOLO v8 to ONNX or TFLITE: (14:13)
---------------------------------------------
► Links:
---------------------------------------------
#TheCodingBug
---------------------------------------------
► My Other Tutorials:
---------------------------------------------
---------------------------------------------
DISCLAIMER: Links included in this description might be affiliate links. If you purchase a product or service with the links I provide, I may receive a small commission. There is no additional charge to you!

[gothicov]

some videos on youtube 1-2 hours, some even 5 hourse but you made simple 15 min full tutorial without many words, thank you very much

[bravoaviator]

simple, easy and straight forward. I found it after 2 months after searching a lot on web. Thank you man

[adityadhekane]

Amazing, I believe this is the best video on yolov8 custom training

[AliOsmanHocam]

Thanks to this video, I finally found the answer to my questions about Yolov8. As of 03.20 on April 20, I started training my dataset consisting of 1700 pictures and 7 classes, I can sleep comfortably now 🥰 Thank you for your useful video.😉

[mncsupply]

ABSOLUT Simple explained, not much usuelles bs. Brother you saved my grade for UNI.
Love from germany, god bless you

[smrmpvk]

What a great video!!! Been looking for this work and no other video was able to show. Thank you very much

[Muzammil-Ahmed]

i cant thank you enough for making this video! This is the first comment i have ever given on youtubea youtube video just because of how good this is! Concise, straight to the point, comprehensive and he even explains everything. For those of you who are wondering if this works in 2024, yes it surely does! 27/01/2024

[tridibroyarjo]

You are the only YouTuber I came to find whose tutorial works. Great job!

[wei]

Great tutorial. Keep up the good work.

[jajefan]

I wish I had watched this tutorial in the beginning! All these other tutorials work with colab and all these other environments but don't mention that pytorch is installed WITHOUT CUDA!!! I had an RTX 4080 and was banging my head against the wall trying to figure out what was wrong when it was the ultralytics package that had installed pytorch and torchvision without cuda, and wondering why the training wasn't happening on my cuda enabled machine.

You also brilliantly showed the package/directory structure that the ultralytics package is expecting, where each directory of datatype (train, val, test) has their own images + labels directories inside, which some other tutorials did not show and just expected you to have.

[MrTakashi]

i have never seen a more concise and precise video on yolov8 anywhere else on youtube. Keep up the good work my man!

[lykagracetabiolo]

very clear and concise tutorial! 👏 please make a video for deployment of trained model on mobile application please!

[omkarprabhu]

This tutorial really helped me, thank you

[ravishkumar]

thanks for this video. in very less time you have explained all most all the things. it helps me a lot. thank you for the usefull video

[ziaurrehman]

I will simply say, you nailed it. thumbs up

[fatanehsadeghi]

this is so usefull thank you so much .excellent and quick explanation. welldone

[paultch]

thanks for showing how simple it is :)

[mayurirakhonde]

Thanks for the excellent explanation. I am having one doubt. What is mAP50 and mAP50-95? What should we measure for accuracy purpose?

[jasonbourn]

Thanks a lot Subscribed precise and practical

[vigneshpadmanabhan]

Amazing tutorial thought. thanks for also using a labeller. this one is too good.