README.md

metadata

title: AAI 521 - ASL Hand Detection
colorFrom: blue
colorTo: green
sdk: gradio
sdk_version: 5.50.0
app_file: app.py
pinned: false
license: mit

AAI 521 - ASL Hand Detection System

American Sign Language hand gesture detection using MediaPipe, Gradio and CNN (TBD).

Project Overview

This application detects and classifies basic static ASL gestures using MediaPipe Hands for hand landmark detection and a rule-based classifier for gesture recognition.

Features:

• Three input modes: Webcam snapshot, Image upload, and Live streaming
• Real-time hand landmark visualization
• Support for 5 ASL gestures (A, V, B, 1, W)
• Production logging for debugging

Setup

# Install UV package manager
curl -LsSf https://astral.sh/uv/install.sh | sh

# Create virtual environment and install dependencies
uv sync

# Run the application
uv run python app.py # (Kill with Ctrl+C to shutdown gracefully in linux environment)

Technical Stack

• Hand Detection: MediaPipe Hands (v0.10.9)
• Classification: Rule-based finger extension analysis
• Frontend: Gradio (with tabbed interface)
• Image Processing: OpenCV, NumPy, CNN (TBD)
• Package Management: UV for reproducible builds
• Logging: Production-ready [INFO]/[WARN] logging

---

Collab Setup (Use Your Own Branch)

# 1. Clone the repository
git clone git@github.com:denjcodes/AAI_521_Final_Project.git
cd AAI_521_Final_Project

# 2. Add HuggingFace Space as a remote
git remote add hf git@hf.co:spaces/d2j666/AAI_521_Final_Project

# 3. Verify remotes
git remote -v

# Create and switch to your branch
git checkout -b <your-branch>

# Pull latest from main
git pull origin main

# Make changes
git add .
git commit -m "Message"

# Push your branch to GitHub
git push origin <your-branch>

# Deploy your branch to HF main (only when ready)
git push hf <your-branch>:main

---

SSH Prerequisites

Before cloning or pushing via SSH: - GitHub -> Settings -> SSH and GPG Keys -> New SSH Key - HuggingFace -> Settings -> SSH Keys -> Add SSH Key