Hand-Gesture-Detection

KO | EN

gitlite — search

#javascript #python #hacktoberfest #typescript #machine-learning #deep-learning #docker #macos #cli #kubernetes #awesome-list #pytorch #ios #open-source

Hand-Gesture-Detection

(★ 10)

This project, Hand Gesture Detection, uses OpenCV and MediaPipe to recognize real-time hand gestures from a webcam feed. It detects landmarks, determines finger states, and identifies various gestures like 'Hi', 'Thumbs Up', 'Victory', 'Fist' etc. The system visually displays the detected gesture on-screen.

Hand-Gesture-Detection 최신버젼 다운로드

최종 버전 다운로드 (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Hand Gesture Detection

## Project Overview
This project implements a real-time hand gesture detection system using computer vision techniques. It leverages MediaPipe for hand landmark detection, OpenCV for video processing, and PyAutoGUI for potential system control based on detected gestures. The system recognizes various hand gestures such as "Hi", "Thumbs Up", "Thumbs Down", "Victory", "Peace", and custom gestures, displaying the detected gesture on the video feed with a semi-transparent text overlay.

## Features
- **Real-time Hand Detection**: Uses MediaPipe's hand tracking to detect and track hand landmarks.
- **Gesture Recognition**: Identifies predefined gestures based on finger positions and thumb orientation.
- **Visual Feedback**: Displays a bounding box around the detected hand and overlays the recognized gesture text.
- **Custom Gestures**: Includes unique gestures like "Maara Jayega...," "Nai niklega jaala," and "Nature's Call !!!".
- **Webcam Support**: Configurable for different webcam indices with error handling and debugging logs.

## Requirements
To run this project, ensure you have the following dependencies installed:
- Python 3.8+
- OpenCV (`opencv-python`)
- MediaPipe (`mediapipe`)
- PyAutoGUI (`pyautogui`)
- TensorFlow (used for potential model integration, though not explicitly in the provided code)
- NumPy (implicitly used by MediaPipe and OpenCV)

Install the dependencies using pip:
```bash
pip install opencv-python mediapipe pyautogui tensorflow
```

## Installation
1. Clone or download this repository to your local machine.
2. Ensure you have Python 3.8 or higher installed.
3. Install the required dependencies using the command above.
4. Connect a webcam to your computer.

## Usage
1. Navigate to the project directory containing the script `Project 4 - Hand Gesture Detection using OpenCV.py`.
2. Run the script using Python:
   ```bash
   python "Project 4 - Hand Gesture Detection using OpenCV.py"
   ```
3. The webcam feed will open, displaying detected hand gestures in real-time.
4. Perform hand gestures in front of the webcam to see the recognized gesture displayed on the screen.
5. Press the `ESC` key to exit the application.
6. You can just simply close the terminal to elimate the code.

## Gesture Definitions
The system recognizes the following gestures based on finger states:
- **Hi**: All fingers up (open hand).
- **Thumbs Up**: Thumb up, other fingers down, thumb not tilted downward.
- **Thumbs Down**: Thumb up, other fingers down, thumb tilted downward.
- **Live Long**: Thumb, index, and middle fingers up, others down.
- **Victory**: Index and middle fingers up, others down.
- **Peace**: Ring and pinky fingers up, others down.
- **Point Up**: Index finger up, others down.
- **Fist**: All fingers down.
- **Maara Jayega...**: Thumb and pinky up, others down.
- **Nai niklega jaala**: Thumb, index, and pinky up, others down.
- **Nature's Call !!!**: Pinky up, others down.

## Code Structure
- **File**: `Project 4 - Hand Gesture Detection using OpenCV.py`
- **Key Functions**:
  - `get_finger_states`: Determines which fingers are up or down based on landmark positions.
  - `is_thumbs_down`: Checks thumb orientation to differentiate between "Thumbs Up" and "Thumbs Down."
  - `detect_gesture`: Maps finger states to specific gestures.
- **Main Loop**:
  - Captures webcam frames.
  - Processes frames with MediaPipe to detect hand landmarks.
  - Draws landmarks and bounding boxes using OpenCV.
  - Displays gesture text with a semi-transparent background.
  - Handles webcam errors and provides debugging logs.

## Notes
- The script uses `CAP_DSHOW` for Windows webcam access. For macOS or Linux, remove or replace `cv2.CAP_DSHOW` with the appropriate backend (e.g., `cv2.CAP_ANY`).
- If the webcam fails to open, the script attempts to use an alternative index (0 or 1). Adjust the index if needed for your system.
- TensorFlow is listed as a dependency but not explicitly used in the provided code. It may be intended for future model-based gesture recognition.
- Ensure sufficient lighting and a clear view of the hand for accurate detection.

## Troubleshooting
- **Webcam Issues**: If the webcam doesn't open, check the connection or try a different index (e.g., `cv2.VideoCapture(2)`).
- **Performance**: Reduce the frame resolution (e.g., 320x240) for faster processing on low-end devices.
- **Gesture Misrecognition**: Ensure the hand is fully visible and within the webcam's field of view.

## Future Improvements
- Integrate TensorFlow for advanced gesture classification using a trained model.
- Add support for multi-hand detection.
- Implement gesture-based system controls using PyAutoGUI (e.g., mouse clicks or keyboard inputs).
- Enhance the UI with more detailed visual feedback or a graphical interface.

Hand-Gesture-Detection

Hand-Gesture-Detection 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Hand-Gesture-Detection

Hand-Gesture-Detection 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Download README