MOFIT Demo

Overview

Simple little demo for the Seminar 490 class. The Python script uses pose estimation to analyze and provide feedback on hammer curls. It detects keypoints on the human body and calculates angle of elbow to determine of each curl. It also counts the number of reps.

Dependencies

• Python 3.x
• OpenCV
• YOLOv8

Use the requirements.txt to install the needed libraries

pip install -r requirements.txt

In addition, you will need to download the Yolov8 Medium Pose Detection Model and put it in the same directory. You can download the offical model from Ultralytics here

Running the Script

Execute the following

python script.py

Note

If you run this script inside VSCode, VSCode may prompt you to allow permission to access your webcam. If you would rather analyze a pre-recorded video than live streaming from your webcam, simply change the VideoCapture(0) source from 0 to the name of your video file.