Machine learning to determine facial measurements via captured images

One or more non-transitory computer-readable media collectively or individually comprising computer-executable instructions that, when executed by one or more processors of one or more processing systems, cause the one or more processing systems to collectively or individually perform an operation comprising:

• identifying a set of images, from a plurality of images, that satisfy defined orientation criteria, comprising, for each respective image in the set of images, determining a respective orientation of a face of a user depicted in the respective image based on a respective set of coordinate locations for a set of facial landmarks, wherein at least one image of the plurality of images is discarded for failing to satisfy the defined orientation criteria;
• estimating a reference distance on the face of the user based on a first image of the set of images, wherein the reference distance indicates a measurement of at least one of the set of facial landmarks;
• estimating a nose depth of the user based on a second image of the set of images based at least in part on the reference distance;
• and
• selecting a facial mask for the user based at least in part on the nose depth.

A method, comprising:

• receiving a first exemplar image, wherein the first exemplar image depicts a face of a user;
• defining one or more regions of interest (ROIs) on the face;
• determining whether the user is wearing glasses in the first exemplar image by processing each of the one or more ROIs using one or more edge detection techniques;
• labeling the first exemplar image to indicate whether the user is wearing glasses;
• and
• refining a machine learning model based on the first exemplar image and label.