README.md

metadata

title: Face Recognition Model Demo App
emoji: 🚀
colorFrom: red
colorTo: red
sdk: docker
app_port: 8501
tags:
  - streamlit
pinned: false
short_description: Face Recognition Project using Transfer Learning( FaceNet em
license: mit

Face Recognition Model Demo App

A fully interactive Face Recognition System built using:

• FaceNet (InceptionResnetV1) for embeddings
• SVM classifier for identity prediction
• Trained on a celebrity face dataset (~18K images, 105 classes)

This Space demonstrates how to deploy a deep-learning + classical ML pipeline using:

• A separate model repository
• A separate dataset repository
• A fast and interactive Streamlit UI
• HuggingFace Hub + Xet for large file handling

🚀 Live Apps

Stable Streamlit App (Recommended)

https://facerecognition-tq32v5qkt4ltslejzwymw8.streamlit.app/

Hugging Face Space (this repo)

https://huggingface.co/spaces/AI-Solutions-KK/face_recognition_model_demo_app

🧩 Architecture Overview

1. Model repo → trained SVM + centroids + classes
   AI-Solutions-KK/face_recognition

2. Dataset repo → structured celebrity dataset
   AI-Solutions-KK/face_recognition_dataset

3. App repo (this Space) → Streamlit UI + inference pipeline
   AI-Solutions-KK/face_recognition_model_demo_app

📁 Project Structure

src/
streamlit_app.py
app/
inference.py
config.py
embeddings_cache/
hf_datasets/
requirements.txt
README.md

🧠 Model Details

Hosted at: https://huggingface.co/AI-Solutions-KK/face_recognition

Includes SVM model, classes, centroids.

📦 Dataset Details

Hosted at: https://huggingface.co/datasets/AI-Solutions-KK/face_recognition_dataset

🖥 Run Locally

git clone https://huggingface.co/spaces/AI-Solutions-KK/face_recognition_model_demo_app
cd face_recognition_model_demo_app
pip install -r requirements.txt
streamlit run src/streamlit_app.py

📚 Full Documentation

https://github.com/AI-Solutions-KK/face_recognition_cnn_svm

👤 Author

Developed by Karan (AI-Solutions-KK)