Hello trend,
This is Satya Mallick from LearnOpenCV.com.
Face Recognition is the computer vision technique that predicts the identity of the person. In today's post you will learn how to make your Face Recognition system more robust agains spoofing attacks! There are 2 main parts to it:
- Perform Face Detection and Recognition using neural network models
- Perform Real v/s spoof classification using the depth maps obtained from OAK-D
Face Recognition v/s Face Verification
Many people think of Face Recognition and Face Verification as different problems and many think of them as same. The answer is that idea behind both is same, just the application area is different.
Face Verification as the name suggests, tries to authenticate a person. For example, you can unlock your smartphone using your Face, but others can't. It is a 1:1 comparison.
A Face Recognition system ( a.k.a Face Identification ) looks for the person in a database of known people and tries to predict who the person is. It is a one to many comparison. Learn more about the fundamentals of Face Recognition in our earlier blog post on Face Recognition.
Face Recognition + Anti-Spoofing Classifier
In our last post, we learnt how to create complex pipelines for OAK-D for using multiple neural network models on the OAK-D device. We will leverage the learnings from that post to create an anti-spoofing Face Recognition system using the OAK-D.
We will be using a Face Detection model followed by a face recognition model called sphereface. You will learn how to enroll a face in the database and how to authenticate a new face with the database. Finally, we train a classifier using the depth maps obtained from OAK-D to classify if the detected face is real of spoofed. Let's learn more in our blog post.
| Anti-Spoofing Face Recognition using OAK-D |
You can download the code by clicking on the button below
| Download Code (GitHub) |
Introduction to Spatial AI
If you have no idea what Spatial AI is, here is a video to get you started.
After you have gone through the video, please check out other posts in the series. You will learn quite a bit about stereo vision in general even if you do not own an OAK-D.
- Introduction to OAK-D and DepthAI
- Stereo Vision and Depth Estimation using OpenCV AI Kit
- Object Detection with Depth Measurement using pre-trained models with OAK-D
- DepthAI Pipeline Overview: Creating a Complex Pipeline
Cheers!
Satya
Courses / YouTube / Facebook / LinkedIn / Twitter / Instagram
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.