This program is a computer vision-based exercise feedback system made by students in UiA Grimstad. it's purpose is to help people who exercise be more aware of how they work out and how to correct themselves.
- real time pose analysis
- posture feedback
- session log
- mediapipe
- opencv-python
- numpy
- any camera attached to the system
- The LandmarkScript.py has already been executed, if the dataset og images is to be changed then it will need to be executed again.
- The program will only track standard barbell and non-barbell squats.
- To start the program you will need to compile the Exercise.py python file and start the tracking, make sure the program detects the entire body when looking at the frame, so the program is more consistent.