Gesture Detection

A real-time hand gesture detection project using deep learning and computer vision. The system detects hand gestures through a webcam and uses trained models to recognize gestures and perform corresponding actions.

Features

• Real-time hand gesture detection
• Webcam-based input
• LSTM-based gesture recognition
• Trained deep learning models
• Gesture-based action control
• Real-time prediction

Technologies Used

• Python
• TensorFlow / Keras
• OpenCV
• NumPy
• MediaPipe
• LSTM

Project Structure

gesture_detection/
│
├── data/
├── gesture_data/
├── collect_data.py
├── train_model.py
├── train_lstm.py
├── realtime_lstm.py
├── live_detection.py
├── gui_simulation.py
├── train_model.py
├── class_labels.json
├── requirements.txt
└── README.md

Installation

Clone the repository:

git clone https://github.com/PiyushRaj282/gesture_detection.git
cd gesture_detection

Install the required libraries:

pip install -r requirements.txt

Usage

To collect gesture data:

python collect_data.py

To train the model:

python train_model.py

To run real-time gesture detection:

python live_detection.py

Project Workflow

Webcam
   ↓
Hand Detection
   ↓
Feature Extraction
   ↓
LSTM Model
   ↓
Gesture Prediction
   ↓
Action / Output

Applications

• Human-computer interaction
• Touchless controls
• Sign and gesture recognition
• Smart device control
• Computer vision applications