/siamese_faces

Siamese neural network for facial recognition

Primary LanguagePythonMIT LicenseMIT

Siamese Faces

Applied Facial Recognition with Siamese Neural Networks

Overview

Facial recognition is a common task performed in a variety of industries. Companies like Facebook use this tool for identifying faces in uploaded user photos, while Baidu uses this for employee facial recognition.

This task seeks to check contrastive loss between similar images of people.

Table of Contents

  1. Dataset
  2. Preprocessing
  3. Model
  4. Results
  5. Tech Stack
  6. References

Data Set

The data consists of 25 head shots from the Galvanize g49 Data Science Immersive cohort.

Preprocessing

All images were converted from RGB to grayscale. Using OpenCV, the faces within each image are cropped, and saved to a new directory. The images are then resized to 512 x 512.

Finally, all images were mirrored using OpenCV. This is to look at contrastive loss between two photos of the same person.

Model

Results

Tech Stack

  • OpenCV

References