Systems and methods for determining an ear saddle point of a user to produce specifications to fit a wearable apparatus to the user's head

1. A method of determining an ear saddle point of an ear of a subject to produce specifications to fit a wearable apparatus to a head of the subject, the method comprising:

generating a three-dimensional mesh surface representing at least a portion of a surface of the head of the subject;

obtaining a profile image of the ear of the subject;

determining, using a first model, of one or more trained models, a boundary of the ear of the subject in the profile image;

generating, using the first model, or a second model, a map of the ear of the subject indicating confidence levels of probable locations of a two-dimensional ear saddle point in a plane of the profile image;

determining a most probable location of the two-dimensional ear saddle point, and a confidence level of the most probable location, based on the map;

comparing the confidence level of the most probable location to a threshold;

retrieving, if the confidence level of the most probable location is less than the threshold, a location of the ear saddle point determined based on measurements performed on the subject;

performing, if the confidence level of the most probable location is greater than or equal to the threshold:

projecting the two-dimensional ear saddle point onto the three-dimensional mesh surface,

determining a maximum depth of the three-dimensional mesh surface in a defined region around the projected two-dimensional ear saddle point, and

determining the ear saddle point based on the projected two-dimensional ear saddle point and the determined maximum depth; and

generating specifications of a wearable apparatus to fit the wearable apparatus to the head of the subject based on the determined ear saddle point.

2. The method of claim 1 , wherein the wearable apparatus is a wearable heads up display and the generated specifications comprise data regarding size of components of the wearable apparatus, including one or more of:

dimensions of temple support portions;

dimensions of a front frame of a support structure; and

dimensions of a bridge.

3. The method of claim 1 , wherein the three-dimensional mesh surface is generated from images from one or more cameras.

4. The method of claim 1 , wherein the location of the ear saddle point based on measurements performed on the subject is determined by placing a jig on the subject and making calculations based on imagery produced by one or more cameras.

5. The method of claim 1 , further comprising obtaining a profile image of an other ear of the subject and determining an ear saddle point location of the other ear.

6. The method of claim 1 , wherein the maximum depth of the three-dimensional mesh surface in the defined region around the projected two-dimensional ear saddle point is measured along a transverse axis extending in a direction outward from the subject's head so as to be perpendicular to a plane of the profile view camera images.

7. The method of claim 1 , wherein determining the ear saddle point based on the projected two-dimensional ear saddle point and the determined maximum depth comprises replacing a depth coordinate of the projected two-dimensional ear saddle point with a value corresponding to the determined maximum depth of the three-dimensional mesh surface in the defined region around the projected two-dimensional ear saddle point.

8. The method of claim 1 , wherein the first model performs object recognition using a neural network model or computer vision algorithm.

9. The method of claim 1 , further comprising generating a data set of two-dimensional ear saddle points to train at least one of the first model or the second model.

10. The method of claim 9 , wherein the generating of the data set of two-dimensional ear saddle points comprises:

placing a jig having one or more target markers on the head of the subject, the jig having a known geometry which comprises positions of portions of the jig which are in contact with one or more ear saddle points of the subject;

detecting positions of the one or more target markers;

computing three-dimensional ear saddle points based on the detected positions of the one or more target markers and the known geometry of the jig; and

computing two-dimensional ear saddle points by projecting the computed three-dimensional ear saddle points onto a profile plane.

11. A method of determining an ear saddle point of an ear of a subject to produce specifications to fit a wearable apparatus to a head of the subject, the method comprising:

generating a three-dimensional mesh surface representing at least a portion of a surface of the head of the subject;

obtaining an off-center frontal image of the head of the subject;

determining, using a first model, of one or more trained models, a boundary of an ear region of the subject in the off-center frontal image;

determining a reference point on the boundary to tie the boundary to coordinates of the three-dimensional mesh surface;

projecting the boundary onto the three-dimensional mesh surface by projecting the determined reference point of the boundary onto a corresponding reference point on the three-dimensional mesh surface;

determining, using the first model, or a second model, the ear saddle point on the three-dimensional mesh surface within the projected boundary; and

generating specifications of a wearable apparatus to fit the wearable apparatus to the head of the subject based on the determined ear saddle point.

12. The method of claim 11 , further comprising determining a confidence level associated with the determining of the ear saddle point.

13. The method of claim 11 , wherein the reference point is a point on the boundary which is closest to the top and front of the head of the subject.

14. The method of claim 11 , wherein the wearable apparatus is a wearable heads up display and the generated specifications comprise data regarding size of components of the wearable apparatus, including one or more of:

dimensions of temple support portions;

dimensions of a front frame of a support structure; and

dimensions of a bridge.

15. A method of determining an ear saddle point of an ear of a subject to produce specifications to fit a wearable apparatus to a head of the subject, the method comprising:

generating a three-dimensional mesh surface representing at least a portion of a surface of the head of the subject;

dividing the three-dimensional mesh surface along a midsagittal plane to produce at least one profile view of the head of the subject;

determining a location of an ear canal based on depth coordinates of points on the three-dimensional mesh surface in the profile view, the depth coordinates being measured along a transverse axis extending in a direction perpendicular to the midsagittal plane;

determining, using a trained model and the determined location of the ear canal, the ear saddle point on the three-dimensional mesh surface; and

generating specifications of a wearable apparatus to fit the wearable apparatus to the head of the subject based on the determined ear saddle point.

16. The method of claim 15 , wherein the apparatus is a wearable heads up display and the generated specifications comprise data regarding size of components of the wearable apparatus, including one or more of:

dimensions of temple support portions;

dimensions of a front frame of a support structure; and

dimensions of a bridge.