Custom object detection in Python using YOLOv8

In this video, I discuss the YOLOv8 data format and how to train a custom object detection model using Ultralytics YOLOv8. Please note that I will be using the Python API and not the CLI. We will be detecting litter in images of water bodies!

Please subscribe and like the video to help me keep motivated to make awesome videos like this one. :)

Follow me on:

[LWatson]

Wow this is genuinely fantastic tutorial, thanks!

[Athens]

do u have an example if u want to do image augmentation so a flip in an image would someone need also to flip the coordinates of the labels.
welcome back to youtube.

[sanjoetv]

can you teach us how to customized yolov8 landmark for measurement. Modify the architecture of the yolov8 landmark to measure and object size. can you do that?

[gunasekhar]

But my data set contains different dimensions of images. I have just completed the image annotations. But in the video he said that image size is very important. So what can j do now?

[bilalshahid]

Hi, can you please tell how can I determine accuracy, precision and f1 score on my testing data?

[mr.musalman]

Thanks, for the video. How to fix that tensor board issue in Kaggle please explain.
Also I wanna a video about segmentation using YOLO v8.
Thanks again for this video.

[prabaldutta]

How to train yolov8 for custom keypoint detection?

[girrajjangid]

Thanks @abhishek. Just a one question if we implement object detection which is better in MaskRCNN / yolo v7/8? I know MaskRCNN is segmentation model but it also produce bbox I am considering that. Your response will highly appreciable

[YigalBZ]

Thanks for the good video. The title is confusing - custom usually means my own date, yet you use an existing dataset, without showing hot to train my own images (labeling etc.).

[venkateshr]

Sir please we need separate book dedicated for transformers 🙏