https://github.com/yareva/gestutext
Real-time hand gesture recognition using MediaPipe and OpenCV, trained with a Random Forest classifier. Detects OK, Yes, No, and Peace Out gestures via camera.
https://github.com/yareva/gestutext
gesture-recognition machine-learning mediapipe opencv python random-forest
Last synced: about 2 months ago
JSON representation
Real-time hand gesture recognition using MediaPipe and OpenCV, trained with a Random Forest classifier. Detects OK, Yes, No, and Peace Out gestures via camera.
- Host: GitHub
- URL: https://github.com/yareva/gestutext
- Owner: yareva
- Created: 2025-01-30T00:26:43.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2025-10-04T02:32:30.000Z (9 months ago)
- Last Synced: 2025-10-11T09:40:14.707Z (9 months ago)
- Topics: gesture-recognition, machine-learning, mediapipe, opencv, python, random-forest
- Language: Python
- Homepage:
- Size: 161 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# GestuText
**GestuText** is a hand gesture recognition system using **MediaPipe**, **OpenCV**, and **scikit-learn**. It detects hand gestures from a camera in real-time and classifies them using a trained Random Forest model.
## 🤖 Supported Gestures
| Gesture | Class | Label |
|----------|--------|-------------|
| 👌 | 0 | OK |
| 👍 | 1 | Yes |
| 👎 | 2 | No |
| ✌️ | 3 | Peace Out |
## 📦 Project Structure
- `collect_data.py` – Captures webcam images of each gesture
- `process_data.py` – Extracts hand landmarks using MediaPipe and saves data to `data.pickle`
- `train_model.py` – Trains a Random Forest classifier and saves `model.p`
- `run_model_live.py` – Uses webcam to detect gestures in real-time and display predictions
## 🚀 How to Run
1. **Collect Data**
Run `collect_data.py` and follow on-screen instructions to capture 200 images per gesture.
2. **Process Data**
```bash
python process_data.py