🧍‍♂️ Posture Analyzer (OpenCV + MediaPipe + Streamlit + Docker)

This project is a Posture Analysis System built using OpenCV, MediaPipe, and Streamlit, capable of tracking knee movement, drawing pose landmarks, and visualizing joint motion paths.
The entire application is Dockerized, making it extremely easy to run anywhere without dependency issues.

---

🚀 Features

• Real-time posture estimation using MediaPipe Pose
• Knee tracking with live plotted trajectories
• Visual output of pose skeleton
• Streamlit interface for:
  • Uploading custom videos
  • Using built-in sample videos
  • Running analysis with a single click
• Fully containerized using Docker
• Works consistently across systems due to Docker isolation

---

📁 Project Structure

PostureAnalysis/
│── Analyzer.py
│── app.py
│── Sample1.mp4
│── Sample2.mp4
│── requirements.txt
│── Dockerfile
└── README.md

---

🧠 How It Works

• The Analyzer.py file handles all processing:
  • Reads video frames
  • Extracts pose landmarks
  • Tracks knee positions
  • Displays pose and extracted keypoints
• The Streamlit app provides a simple UI for interacting with the analyzer.
• With Docker, everything runs the same on any machine.

---

🖥️ Running Locally (Without Docker)

1️⃣ Create a virtual environment

python -m venv venv

2️⃣ Activate it

Windows:

venv\Scripts\activate

Mac/Linux:

source venv/bin/activate

3️⃣ Install dependencies

pip install -r requirements.txt

4️⃣ Run Streamlit app

streamlit run app.py

---

🐳 Running Using Docker (Recommended)

1️⃣ Build Docker Image

Run this inside your project folder:

docker build -t marti18/analyzer .

---

2️⃣ Run the Container

docker run -p 8501:8501 marti18/analyzer

Now open your browser and go to:

👉 http://localhost:8501

---

3️⃣ Push Image to Docker Hub

Login

docker login

Tag image

docker tag marti18/analyzer marti18/analyzer:latest

Push image

docker push marti18/analyzer:latest

You can now pull it from anywhere using:

docker pull marti18/analyzer
docker run -p 8501:8501 marti18/analyzer

---

🧪 Example Usage

▶ Run analysis on:

• Uploaded video
• Sample1.mp4
• Sample2.mp4

Streamlit provides:

• Pose skeleton window
• Extracted pose window
• Knee trajectory plot

---

📦 Requirements

All dependencies are listed inside requirements.txt:

opencv-python
mediapipe
numpy
matplotlib
streamlit

---

🛠 Technologies Used

• Python
• OpenCV
• MediaPipe
• Matplotlib
• Streamlit
• Docker

---

✨ Author

Aditya Pratap Singh