Suppose you are a Machine Learning Engineer in a company. One day a robbery happens in your company's headquarters. So to improve security, the security team decides to permit entry only to the employees inside the headquarters.
You are assigned a task to build a machine learning model for detecting whether a person is an employee of the company or not using facial recognition. You are given a dataset which contains photos of all employees.
Now the model's task will be to scan the face of a person and predict whether the person is an employee of the company or not. This task can be performed using-
Our approach - While training the model, we have a dataset consisting of photos of employees. So, we convert all the images into encodings.
Then when it comes to prediction, we scan the face of the person and convert it into encoding and then match encoding with the stored encodings. When we find much similarity between two images i.e. similarity above a certain pre-decided threshold, then we can say that the person is an employee.
If we don't find any encoding with similarity greater than the threshold, we can say that the person is not an employee.
Taking you to the next exercise in seconds...
Want to create exercises like this yourself? Click here.
Note - Having trouble with the assessment engine? Follow the steps listed here
Loading comments...