Hello everyone,
I have a question regarding the annotation process for a keypoint detection project. I am not sure whether I am annotating my dataset correctly. I have searched through different communities but could not find anyone with the same problem, and most YouTube tutorials only explain how to perform training, not how to annotate.
Here are the details of my setup:
-
I defined the class for my dataset
-
I placed the keypoints on top of a tennis court image
-
I prepared a training dataset with multiple annotated images
-
I trained a YOLOv11 model and checked the predictions
However, the results are always wrong. I have already tried with around 50 images and even extended it to 300 images, but the predictions are still not working properly.
My main concern is:
-
Is my annotation method incorrect?
-
Or is there something else I am missing to make the keypoints detectable by YOLOv11?
Any help, guidance, or example of correct annotation practices for YOLO keypoint projects would be greatly appreciated.
Thank you in advance!
- Project Type: Keypoint Detection
- Operating System & Browser: Windows & Google Chrome
- Project ID: tennis-court-suuzy
