Nov 2025
Burpee Exercise Tracker
Top: screenshot of setup screen. Middle: screenshot of Burpee Exercise Tracker while testing pose estimation and video annotation. Bottom: screenshot of Burpee Exercise Tracker when actually doing burpees.
Burpees are one of my favorite exercises, but I often lose count when doing 100 or more consecutively…
To solve this, I developed the Burpee Exercise Tracker. This web app uses computer vision to estimate and track the user’s body position, automatically counting completed burpees. It also provides real-time metrics like total time, burpee speed, and current position.
The project combines a Python backend (OpenCV and MediaPipe) with an HTML/CSS/JavaScript frontend via Flask. Building it taught me a lot about pose estimation, program flow control, and integrating Python with web interfaces.
In the future, I’m excited about applying pose estimation techniques from this project to my work in prosthetic devices, helping to better understand and predict human movement for more intuitive and reliable prostheses.
Screenshot of the Burpee Exercise Tracker while testing pose estimation and video annotation. MediaPipe is used to locate landmarks on the user’s body (shoulders, elbows, wrists, hips, knees, ankles). OpenCV is used to annotate the real-time video stream shown in the web app interface.