Methods and systems herein can include a processor configured to obtain a set of images that shows a user's head and a reference object, generate a user's head model of the user's head based at least in part on the set of images, and generate a reference object model of the reference object based at least in part on the set of images. The processor can further be configured to determine an orientation and a size of the reference object model based at least in part on a relative location of the reference object relative to the user's head in the set of images and use the reference object model, the orientation of the reference object model, the size of the reference object, and a known dimension of the reference object to determine scaling information. The processor can then be configured to apply the scaling information to the user's head model to obtain a scaled user's head model. The system can also include a memory coupled to the processor and configured to provide the processor with instructions.