https://github.com/zolppy/celebrity-face-recognition
https://github.com/zolppy/celebrity-face-recognition
cnn computer-vision deep-learning face-detection kaggle keras machine-learning object-classification opencv python tensorflow
Last synced: 3 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/zolppy/celebrity-face-recognition
- Owner: zolppy
- License: mit
- Created: 2025-08-22T00:40:04.000Z (11 months ago)
- Default Branch: main
- Last Pushed: 2025-08-22T01:24:22.000Z (11 months ago)
- Last Synced: 2025-08-22T02:35:37.226Z (11 months ago)
- Topics: cnn, computer-vision, deep-learning, face-detection, kaggle, keras, machine-learning, object-classification, opencv, python, tensorflow
- Language: Jupyter Notebook
- Homepage:
- Size: 12.7 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Celebrity Face Recognition
Celebrity face recognition system built using deep learning. It leverages a Convolutional Neural Network to identify individuals from a dataset of celebrity images. The system first pre-processes the image data, trains the model, and then uses a Haar Cascade classifier to detect faces in new images. Finally, it uses the trained model to predict and display the identity of the detected celebrity.
---
## 📂 Project Structure
├── main.ipynb # Jupyter Notebook with code, training, and evaluation
├── README.md # Project documentation
---
## 📊 Dataset
The dataset is downloaded using [`kagglehub`](https://pypi.org/project/kagglehub/).
It contains labeled images across multiple categories. Images are preprocessed using **OpenCV** and split into training and testing sets.
---
## 🧠 Model Architecture
The CNN is implemented using **TensorFlow/Keras** with the following structure:
- **Conv2D** + **MaxPooling2D** layers for feature extraction
- **Flatten** layer to convert feature maps into vectors
- **Dense layers** with ReLU activations
- **Softmax output layer** for classification
The model is compiled with:
- Optimizer: `Adam`
- Loss: `categorical_crossentropy`
- Metrics: `accuracy`
---
## ⚙️ Installation
Clone this repository and install the dependencies:
```bash
git clone https://github.com/yourusername/celebrity-face-recognition.git
cd celebrity-face-recognition
pip install -r requirements.txt
```
## Requirements
- Python 3.8+
- TensorFlow
- Keras
- OpenCV
- NumPy
- Matplotlib
- scikit-learn
- kagglehub
You can install them manually with:
```bash
pip install tensorflow keras opencv-python numpy matplotlib scikit-learn kagglehub
```
---
## 🚀 Usage
Run the notebook step by step:
```
jupyter notebook main.ipynb
```
The notebook includes:
- Dataset download & preprocessing
- Model training
- Evaluation & accuracy results
- Visualization of sample predictions
---
## 📈 Results
The model achieves competitive accuracy on the test dataset.
Example prediction visualization:
- Input image shown with label
- Model prediction displayed with confidence
(See `main.ipynb` for detailed plots and accuracy results.)
---
## 📌 Future Work
- Improve performance with data augmentation
- Experiment with deeper CNN architectures
- Deploy the trained model with a web interface
---
## 🤝 Contributing
Pull requests are welcome. For major changes, please open an issue first to discuss what you’d like to modify.
---
## 📜 License
This project is licensed under the MIT License.