This project delves into the fascinating realm of facial recognition and machine learning to create a fun and interactive application. By leveraging pre-trained deep learning models and image processing techniques, users can discover which celebrity they resemble most based on their facial features!
Key Features: Celebrity Image Recognition: Employs a pre-trained convolutional neural network (CNN) to extract features from celebrity images, creating a robust reference database. User Image Processing: Guides users through uploading their photo, which is then preprocessed for compatibility with the CNN model. Similarity Calculation: Computes the cosine similarity between the user's facial features and those of celebrities in the database, identifying the closest match. Interactive Results: Presents the celebrity with the highest similarity score, potentially sparking amusement or surprise!
Technology Stack: TensorFlow/Keras: Provides the foundation for deep learning model implementation and execution. Scikit-learn: Offers functionalities for calculating cosine similarity or other distance metrics, if applicable. NumPy: Handles numerical computations and array manipulations. Pickle: Employed for efficiently storing and loading pre-processed data (e.g., celebrity features).
Usage: Clone the repository: Run git clone https://github.com/AgarwalBhavya/Which-Celebrity-Are-You.git. Activate virtual environment. Run the script: Execute python test.py (or the appropriate script name). Follow on-screen instructions: The script might prompt you to upload an image or provide the path to your pre-processed data files.
.png?Expires=1776223855&Key-Pair-Id=K2V2TN6YBJQHTG&Signature=gxtu11I3snViOEEdpKnx24GuvdrdFlpNRjJyYSKURWvQKn~RavosMbGRH21L1i5PTqzNl2M6P7IYmQc19zyxSMtC3Xiip3I7Y09-AW6mqvk2LjINBJEPZFURddhXgSnyExfAYqQow4SGNDFL03KUiJKLnhOvXI9H8vKjmaAMXCLvLh4beqkUm231Dxt8TwHiPSnD0chEzzj8iT9NVRFv-1Y8OQD7201692sxYl1PSqHaAMuiFkm8GacxWsv1caAu0GuRCROFwqVshT0tRW86DOxYRBWjpPDGiIT4a9c1MCDJXKWn62WXrn1TvEmqBN0mcqUXakj5VKPg~-ZjZB1EZg__)