Project 4: Face Recognition | AI_WQ

Overview

Face detection and recognition using MTCNN for face detection and InceptionResNet for embedding extraction. The project includes deployment as a Flask web API.

MTCNN InceptionResNet Embeddings Flask API Face Images

Notebooks

Notebook	Topic	Description
043	MTCNN	Multi-task Cascaded CNN for face detection
044	InceptionResNet	Face embedding extraction for recognition
045	Flask API	Web service deployment for face recognition

Python Modules

Module	Description
`facenet.py`	FaceNet embedding model implementation
`wq_face_recognition.py`	Complete face recognition pipeline
`wq_app.py`	Flask web application for API deployment

Key Techniques

MTCNN Detection - Multi-stage face detection with bounding boxes and landmarks
Face Embeddings - 512-dimensional vectors for face representation
Similarity Matching - Cosine similarity for face comparison
API Deployment - RESTful endpoints for face detection and recognition

Available Tools

Tool	Description	Variation
`load_proj4_info`	Get Project 4 information and available resources	V2+
`list_proj4_modules`	List face recognition Python modules	V2+