From the course: Deep Learning: Face Recognition
Join today to access over 22,600 courses taught by industry experts or purchase this course individually.
Identifying a face from face encodings - Python Tutorial
From the course: Deep Learning: Face Recognition
Identifying a face from face encodings
- [Instructor] Using our face encoding model we can feed in the picture of a face and get back 128 measurements to represent that face. So how can we use those measurements to identify people? We'll do that using Euclidean distance. Euclidean distance is just the distance between two points in space along a straight line. It can be hard to visualize a model with 128 measurements. To make things easier to visualize, let's use a much simpler example. We'll manually take three measurements from both of these faces. For the face on the left, we'll say the length of the nose is 2.5, the length of the mouth is 2.5, and the distance between the eyes is four. For the face on the right, we'll say the length of the nose is two, the width of the mouth is three, and the distance between eyes is 3.5. Now we can visualize those sets of measurements as two points in 3D space. The face on the left is at 2.5, 2.5, 4, shown here as a point in blue, and the face on the right is at 2, 3, 3.5, shown here…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.