https://github.com/rayfin774/face-and-hand-gestures-detection-model

Real-time Face & Hand Detection using MediaPipe Framework which Detects emotions (Happy, Sad, Angry) and Gestures (Thumbs Up, Peace, Pointing) live via webcam!
https://github.com/rayfin774/face-and-hand-gestures-detection-model

matplotlib mediapipe opencv python sckiit-learn

Last synced: 8 months ago
JSON representation

Real-time Face & Hand Detection using MediaPipe Framework which Detects emotions (Happy, Sad, Angry) and Gestures (Thumbs Up, Peace, Pointing) live via webcam!

• Host: GitHub
• URL: https://github.com/rayfin774/face-and-hand-gestures-detection-model
• Owner: rayfin774
• Created: 2025-11-07T13:06:05.000Z (9 months ago)
• Default Branch: main
• Last Pushed: 2025-11-07T14:48:55.000Z (9 months ago)
• Last Synced: 2025-11-07T15:19:00.833Z (9 months ago)
• Topics: matplotlib, mediapipe, opencv, python, sckiit-learn
• Homepage:
• Size: 5.64 MB
• Stars: 1
• Watchers: 0
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
 ## Facial Emotion and Hand Gesture Detection using MediaPipe & Machine Learning

##  Overview of my Project

This project presents a **real-time Facial Emotion and Hand Gesture Detection System** that leverages **MediaPipe** and **Machine Learning** to identify human facial expressions and hand gestures using a live webcam feed.  

The system aims to enhance **human–computer interaction** by enabling intuitive communication through **facial emotions** and **hand gestures**.

It can detect and classify multiple expressions such as *happy, sad, angry*, and recognize gestures like *thumbs up, peace, pointing,* and *open hand* in real-time.

---

##  Key Features

-  **Real-Time Detection:** Face and hand landmarks captured via webcam.  

-  **Emotion Recognition:** Detects happiness, sadness, and anger based on facial landmarks.  

-  **Gesture Recognition:** Recognizes common hand gestures like thumbs up, peace, etc.  

-  **Machine Learning Integration:** Uses a trained Random Forest model for classification.  

-  **Interactive GUI:** Displays live predictions for user engagement.  

-  **Model Serialization:** Models saved using Joblib for reusability.  

---

##  Tools & Technologies 

| Category | Tools Used |

|-----------|-------------|

| **Programming Languages** | Python |

| **Libraries & Frameworks** | MediaPipe, OpenCV, Scikit-learn, Joblib, Matplotlib |

| **Hardware Requirement** | Webcam-enabled device |

| **IDE / Environment** | VS Code, Jupyter Notebook, XAMPP (for web integration) |

---

##  Project Workflow

1. **Data Collection & Preprocessing**  

   - MediaPipe detects facial (468 points) and hand (21 points) landmarks.  

   - Extracted coordinates converted into numerical feature vectors.  

2. **Model Training**  

   - Random Forest Classifier trained on processed facial and gesture data.  

   - Models saved using Joblib (`.pkl` format) which is converted to (`.csv` format) .  

3. **Real-Time Detection**  

   - Live video feed analyzed using OpenCV.  

   - Facial expressions and gestures predicted and displayed on GUI.  

---

###  Clone the Repository

```bash

git clone https://github.com//Facial-Emotion-and-Hand-Gesture-Detection.git

cd Facial-Emotion-and-Hand-Gesture-Detection