Use YoloV8 pose detection to get a human keypoint and save it to a CSV file for training a Machine learning and Neural Network for detecting human pose, In this section I will detect if the human is in a cutting pose or not.

---

Before you run the code, you must download the yolov8 keypoint detection weight in this link : https://docs.ultralytics.com/tasks/pose/#models You can choose whatever you want, but I recommend the YOLOv8n-pose because it has a lower computation than the others.

If you just want to run the model, you can download my DNN/CNN/Xgboost weight via this link : https://drive.google.com/drive/folders/1ZddvgCtaz3Hvq6Xp3teR6VyokVCegpiR?usp=sharing

for CNN and DNN, you must download the JSON file too because the JSON file saves my CNN/DNN architecture.

For the video result of the work, visit this link : https://drive.google.com/drive/folders/15A0Btp7llxlFWEHZBADPnHOoAOsgmNEd?usp=sharing

---

Reference :

1. https://alimustoofaa.medium.com/yolov8-pose-estimation-and-pose-keypoint-classification-using-neural-net-pytorch-98469b924525
2. https://github.com/mmakos/HPC
3. https://s3.eu-central-1.amazonaws.com/ucu.edu.ua/wp-content/uploads/sites/8/2020/11/Zakharchenko-Iryna_188601_assignsubmission_file_Iryna-Zakharchenko.pdf