Mask Detection based on Create ML

Project Collaborators：

• Michael
• Alex
• Steven
• Hanson

Overview

Our intention is to build an automated face mask detection model to test whether people wearing masks appropriately. The model can be deployed in iOS system.

Project Design

This project conducts model training based on Create ML.

The App used in the project comes from the Sample Code of Apple developer website named Recognizing Objects in Live Capture.

The data set comes from Kaggle and includes 853 images.

The dataset is divided into three classes:

• Wearing Mask
• Not Wearing Mask
• Wearing Mask Improperly

Dataset are Labeled by using IBM's Cloud Annotations feature.

The trained model can be deplyed in iOS system using Xcode.

Detailed explanation of this project is in this Video.