Face Detection and Pose analysis

About the Project

This project is a Real-Time Facial & Pose Analysis System designed to monitor live video streams or recordings and provide actionable insights.

It combines facial recognition, emotion detection, and pose estimation to track individuals, log their demographics, and identify actions such as standing or raising hands.

Key features include:

• Face Detection & Tracking: Uses OpenCV’s CSRT tracker for smooth real-time tracking
  
• Face Recognition & Embeddings: DeepFace (Facenet) to uniquely identify individuals
  
• Emotion & Demographics Analysis: Detects age, gender, race, and top emotions
  
• Pose Detection: MediaPipe Pose to classify simple actions
  
• Threaded Processing: Multi-threaded architecture for real-time performance
  
• Unique Face Database: Logs unique individuals with IDs for further analysis
  

This system can be used for student monitoring, workplace safety, or general surveillance, providing insights in real-time while preserving performance.

[WIP will update the about me with my real life problems and experiences at a later time]

Built With

• deepface
• mediapipe
• numpy
• opencv
• pytorch
• tensorflow