Welcome to this tutorial on installing and setting up DeepFace for biometric analysis. In this video, we will walk you through each step, from cloning the repository to running a test script that processes images and outputs detailed JSON files.

First, we start by opening the terminal and cloning the DeepFace repository from GitHub. You will see the command „git clone deepface“ being executed. This downloads all the source files to your local machine, giving you direct access to the code.

Next, we install DeepFace using pip in editable mode with the command „pip install -e“. Editable mode lets you make changes to the code and see those changes immediately without reinstalling. This is very useful for development or customization.

Once DeepFace is installed, we create two folders: one named „inputs“ for the images you want to analyze, and another named „outputs“ where the results will be saved. These folders help keep everything organized.

After setting up the directories, we create a file called „test.py“. In this file, we write a Python script that automatically reads every image in the „inputs“ folder. The script uses DeepFace to analyze each image and extract detailed biometric data. It produces a JSON output that includes key information such as the estimated age, detected emotions, gender, race, and even the face region coordinates. Additionally, the script extracts the facial embeddings—these are long vectors of numbers that uniquely represent the features of each face.

In the final part of the video, we run the script and show two examples: one output for a woman and one for a man. The JSON files clearly display all the biometric details. For instance, you can see the estimated age, the dominant emotion like happy or neutral, and the confidence level of the face detection. The embeddings section provides a high-dimensional numerical signature that can later be used for deep learning tasks such as matching or clustering faces.

This workflow—from cloning the repository, installing in editable mode, creating the necessary directories, writing the test script, to running the analysis—demonstrates an effective way to set up DeepFace. It not only shows you how to extract detailed facial data but also sets the stage for future projects where you might match these embeddings with new attributes in a custom application.

Thank you for watching, and we hope this guide helps you get started with DeepFace and biometric analysis.