Our Digital Products for occupancy counting, social distancing and face mask detection combine video with AI-powered analytics to detect and analyze statistical patterns on how people are complying with policies around max occupancy rates, social distancing and the use of face masks. Providence Health care works closely with multidisciplinary teams at University of British Columbia (UBC), specifically the Department of Computing Science and the Computer Vision Lab. The model is available in public beta subject to further evaluation and training. This model is not intended for medical use.

This machine learning model uses computer vision to detect aggregate level of face mask use in waiting rooms within hospital waiting rooms. The model is currently in the pilot phase at St. Paul's Hospital in Vancouver, Canada in order to ensure physical distancing and compliance to infection control procedures for the safety of our patients, staff and visitors. The model utilizes SSD (Single Shot Detection) objection detection model and weights by applying a single neural network to the full image. No personally identifiable information is stored as part of the analysis.

SSD has two components: a backbone model and SSD head. Backbone model is a pre-trained image classification network as a feature extractor. The backbone in this model has 8 conv layers with 1.01M parameters. This model can extract semantic meaning from the input image while preserving the spatial structure of the image albeit at a lower resolution. Input size of the model is 260x260. The total model has only 24 layers with the location and classification layers counted. Code was inspired by https://github.com/AIZOOTech/FaceMaskDetection