Skip to content

GeekAvi007/PostureSence-MCU-API

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

🧠 PostureSense (Demo Version)

PostureSense is an ultra-lightweight, real-time posture & activity detection API, optimized for microcontrollers and edge devices like ESP32 and Raspberry Pi.
This public version includes a demo simulation of posture output — the full classification logic is private for security, licensing, and IP protection.


🚀 Overview

PostureSense brings intelligent posture monitoring to embedded systems without relying on the cloud or high-performance GPUs. Designed for real-time, offline use, it's perfect for integrating into smart chairs, classrooms, and wellness devices.

Built with TinyML and Edge AI, PostureSense transforms posture monitoring from bulky models to minimal, firmware-friendly APIs — designed for real-world deployment.


🔍 Key Features

Feature Description
🧠 Posture Detection Detects slouching, head tilt, hand raise, and more
⚡ Offline Inference Runs entirely on-device (no internet needed)
🔌 Edge-Ready API Modular Flask-based API for integration with microcontrollers
📡 Serial / MQTT Output Easy communication with embedded firmware
🔒 Privacy-Friendly No cloud, no data upload — all processing happens locally
💼 Monetization-Ready SDK licensing, analytics dashboard, OEM integration roadmap

🧪 Demo Instructions

▶️ Run Locally

git clone https://github.com/GeekAvi007/PostureSence-MCU-API
cd posturesense-demo
pip install -r requirements.txt
python api/server.py
Visit: http://127.0.0.1:5000/posture
You'll receive simulated JSON output mimicking posture predictions.

📦 Output Format (Simulated)
json
Copy
Edit
{
  "posture": "slouch",
  "attention": false,
  "confidence": 0.85,
  "note": "Simulated output. Contact for full access."
}
📚 Use Cases
🪑 Smart Chair
Triggers buzzer or LED if poor posture is detected for more than 3 minutes.

🎓 Student Focus API
Flags inattentiveness in live online classrooms and sends alerts to dashboards.

👴 Elderly Monitor
Detects slouching or unusual head positions and alerts caregivers locally.

🧑‍💻 DevPosture Tracker
Rates your coding posture over time and shows analytics in a dashboard.

🧠 Why It’s Different
🧩 Built for Edge, Not the Cloud
Most AI APIs require internet, servers, or cloud inferencing. PostureSense is optimized for offline usage on microcontrollers — from Raspberry Pi Zero to ESP32-CAM.

📦 You Built an API, Not Just a Model
This isn't just ML — it's a modular developer-ready API designed with:

Licensing potential

Real-world device integration

SDK & dashboard bundling

🔬 TinyML + Embedded Deployment = Rare Combo
Very few developers specialize in:

TinyML optimization

Real-time posture/gesture AI

Deployable edge APIs for hardware startups

You're building what < 0.01% of devs even attempt.

🔐 Want Full Access?
This is a demo-only repository.
The real-time posture classifier, model logic, and SDK are privately maintained.

For full access, licensing, OEM integrations, or pilot collaboration:

📩 Email: avishekmachinelearning@gmail.com
💼 LinkedIn: Avishek Mukherjee

📄 License
rust
Copy
Edit
© 2025 Avishek Mukherjee

You are free to view and use this repository for non-commercial exploration.
Commercial use, redistribution, or reproduction of the underlying classification logic or proprietary modules is strictly prohibited.
Contact the author for licensing or OEM integration.
✅ GitHub Actions (Optional)
If you’ve set up GitHub Actions, you can add this to the top of your README:

markdown
Copy
Edit
![Build Status](https://github.com/GeekAvi007/PostureSence-MCU-API/actions/workflows/python-app.yml/badge.svg)
🎥 Demo GIF
Drop a short GIF here showing a simulated output request:

markdown
Copy
Edit
![PostureSense Demo](static/demo.gif)
🙌 Credits
Built with ❤️ using Flask + OpenCV + TinyML design principles.

“Create products, not just projects.”

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages