PalmLabs

Sign Language Recognition Project (HackMIT post)

PalmLabs combines a Google Chrome Extension and a web application to make communication more accessible through advanced computer vision technology.

---

🚀 Quick Start

Chrome Extension

1. Navigate to extension directory:

   cd extension
   

2. Install dependencies:

   npm install
   

3. Build the extension:

   npm run build
   

4. Load the built extension in Google Chrome for testing

Web Application

Access our web app at: https://localhost:5173/

---

🖥️ Computer Vision

• Custom CNN model trained by RavenOnur
• Model available at: https://huggingface.co/RavenOnur/Sign-Language
• Inference performed using PyTorch

Next Steps

Train our own CNN model including recognition of static words

---

🛠️ Technology Stack

Component	Technologies
Frontend	• React with TypeScript • Convex
Backend	• Node.js • FastAPI
Data Storage	• Convex • PostgreSQL

Special Technologies

• TuneHQ: LLaMa model hosting
• Convex Full-Stack: Seamless frontend-backend integration
• Clerk Auth: Secure user authentication

---