Video 0-24 secs: Posture Analysis | 24 secs + squat counting
🏋️ Real-Time Pose Estimation Fitness & Posture App
A practical computer vision project that tracks body posture and movement in real time. It uses pose estimation to detect key body joints, count exercise repetitions, and assess the quality of movements such as squats, gym exercises, stretching, and yoga poses.
Designed for fitness monitoring and posture-focused experiments, the system can run from a fixed camera or an iPhone feed. It provides live visual feedback on screen, highlighting landmarks and angles so users can see how their form changes during each exercise.
✨ Key Features
- Monitors posture and body alignment in real time
- Counts squats and other exercise repetitions automatically
- Tracks arms, legs, and joint angles dynamically
- Supports live video from camera or iPhone input
- Displays pose landmarks and overlays directly on screen
🔍 How It Works
Each frame from the camera is processed by a pose estimation model that identifies key joints such as shoulders, hips, knees, and ankles. These landmarks are used to compute joint angles and movement patterns, which drive the rep-counting logic and form assessment, with results shown as real-time overlays on the video feed.
🎯 Practical Uses
This project is useful for home workouts, gym-based training, posture assessment, and demonstrations of applied computer vision in sport and healthcare. It provides an intentionally lightweight but effective example of how pose estimation can be turned into an interactive, feedback-driven fitness tool.