Fit prediction based on detection of metric features in image data

1 . A computer-implemented method, comprising:

generating a three-dimensional model of a head of a user based on depth data extracted from a correlation of a change in position and orientation of a computing device with a change in position and orientation of at least one feature between a first image data and a second image data, the at least one feature including a distance between at least two elements detected at locations fixed in a background area, the first image data and the second image data including the head of the user and captured via an application executing on the computing device; and

predicting, by a machine learning model accessible to the computing device, a fit of a first wearable device of a plurality of wearable devices on the head of the user based on the three-dimensional model of the head of the user;

wherein the predicting includes predicting a fit of the first wearable device according to at least one measurement extracted from the three-dimensional model and fit data indicating a fit of respective ones of the plurality of wearable devices on a plurality of users, the fit data based on a plurality of different measurements of the plurality of users, wherein the at least one measurement extracted from the three-dimensional model corresponds to one or more of the plurality of different measurements.

2 . The computer-implemented method of claim 1 , wherein the at least one feature includes at least two facial landmarks that are representative of a facial measurement.

3 . The computer-implemented method of claim 2 , wherein the facial measurement includes at least one of:

a distance between a first ear saddle point and a second ear saddle point representative of a head width of the user;

a distance between an outer corner portion of a right eye and an outer corner portion of a left eye of the user; or

a distance between an inner corner portion of a right eye and an inner corner portion of a left eye of the user.

4 . The computer-implemented method of claim 1 , wherein the at least one feature includes a plurality of features, including:

at least one facial feature defined by two facial landmarks.

5 . The computer-implemented method of claim 1 , wherein detecting the change in the position and the orientation of the computing device includes:

detecting a first position and a first orientation of the computing device in response to receiving first data provided by an inertial measurement unit of the computing device at a capturing of the first image data;

detecting a second position and a second orientation of the computing device in response to receiving second data provided by the inertial measurement unit of the computing device at a capturing of the second image data; and

determining a magnitude of movement of the computing device corresponding to the change in the position and the orientation of the computing device based on a comparison of the second data and the first data.

6 . The computer-implemented method of claim 5 , wherein correlating the change in the position and the orientation of the computing device with the change in the position of the at least one feature includes:

associating the magnitude of the movement of the computing device to a change in a measurement associated with the at least one feature; and

determining the depth data based on the associating.

7 . The computer-implemented method of claim 1 , further comprising:

repeatedly capturing image data as the computing device is moved relative to the user to capture image data from a plurality of different positions and orientations of the computing device relative to the head of the user;

correlating a plurality of changes in position and orientation of the computing device with a corresponding plurality of changes in position of the at least one feature detected in the image data;

determining depth data as the image data is repeatedly captured from the plurality of different positions and orientations based on the correlating; and

developing the three-dimensional model of the head of the user for predicting the fit of the wearable device based on the repeatedly capturing of the image data by the computing device from the plurality of different positions and orientations and the depth data determined from the repeatedly capturing of the image data.

8 . The computer-implemented method of claim 1 , wherein predicting, by the machine learning model accessible to the computing device, the fit of the wearable device includes:

selecting a wearable device, from the plurality of wearable devices, based on the at least one measurement.

9 . The computer-implemented method of claim 8 , wherein the at least one measurement includes at least one of:

a cranial measurement determined based on distance between two facial features detected in the second image data and the first image data; or

an ophthalmic measurement determined based on a distance between two optical features detected in the second image data and the first image data.

10 . The computer-implemented method of claim 1 , wherein the at least two elements detected in a background area include geometric features detected in a repeating pattern.

11 . The computer-implemented method of claim 1 , further comprising generating a prompt that includes instructions to repeat a capturing of the first image data and the second image data with a change to an ambient environment to include features in a background of the first image data and the second image data.

12 . A non-transitory computer-readable medium storing executable instructions that when executed by at least one processor of a computing device are configured to cause the at least one processor to:

generate a three-dimensional model of a head of a user based on depth data extracted from a correlation of a change in position and orientation of the computing device with a change in position and orientation of at least one feature between a first image data and a second image data, the at least one feature including a distance between at least two elements detected at locations fixed in a background area and the first image data and the second image data including the head of the user captured via an application executing on the computing device; and

predict, by a machine learning model accessible to the computing device, a fit of a first wearable device of a plurality of wearable devices on the head of the user based on the three-dimensional model of the head of the user;

wherein the predict by a machine learning model includes predicting a fit of the first wearable device according to at least one measurement extracted from the three-dimensional model and fit data indicating a fit of respective ones of the plurality of wearable devices on a plurality of users, the fit data based on

a plurality of different measurements of the plurality of users associated with the at least one measurement extracted from the three-dimensional model.

13 . The non-transitory computer-readable medium of claim 12 , wherein the at least one feature includes at least two facial landmarks that are representative of a facial measurement, including at least one of:

a distance between a first ear saddle point and a second ear saddle point representative of a head width of the user;

a distance between an outer corner portion of a right eye and an outer corner portion of a left eye of the user;

a distance between an inner corner portion of the right eye and an inner corner portion of the left eye of the user; or

a distance between a pupil of the right eye and a pupil of the left eye of the user.

14 . The non-transitory computer-readable medium of claim 12 , wherein the executable instructions cause the at least one processor to detect the change in the position and the orientation of the computing device, including:

detect a first position and a first orientation of the computing device in response to receiving first data provided by an inertial measurement unit of the computing device at a capture of the first image data;

detect a second position and a second orientation of the computing device in response to receiving second data provided by the inertial measurement unit of the computing device at a capture of the second image data; and

determine a magnitude of movement of the computing device corresponding to the change in the position and the orientation of the computing device based on a comparison of the second data and the first data.

15 . The non-transitory computer-readable medium of claim 14 , wherein the executable instructions cause the at least one processor to:

associate the magnitude of the movement of the computing device to a change in a measurement associated with the at least one feature; and

determine depth data based on the associating.

16 . The non-transitory computer-readable medium of claim 12 , wherein the executable instructions cause the at least one processor to:

repeatedly capture image data as the computing device is moved relative to the user to capture image data from a plurality of different positions and orientations of the computing device relative to the head of the user;

correlate a plurality of changes in position and orientation of the computing device with a corresponding plurality of changes in position of the at least one feature detected in the image data;

determine depth data as the image data is repeatedly captured from the plurality of different positions and orientations based on the correlating; and

develop the three-dimensional model of the head of the user for predicting the fit of the wearable device based on the repeatedly capturing of the image data by the computing device from the plurality of different positions and orientations and the depth data determined from the repeatedly capturing of the image data.

17 . The non-transitory computer-readable medium of claim 12 , wherein the executable instructions cause the at least one processor to:

select a wearable device, from the plurality of wearable devices, based on the at least one measurement, the at least one measurement including at least one of:

a cranial measurement determined based on distance between two facial features detected in the second image data and the first image data; or

an ophthalmic measurement determined based on a distance between two optical features detected in the second image data and the first image data.

18 . A system, comprising:

a computing device, including:

an image sensor;

at least one processor; and

a memory storing instructions that, when executed by the at least one processor, cause the at least one processor to:

generate a three-dimensional model of a head of a user based on depth data extracted from a correlation of a change in position and orientation of the computing device with a change in position and orientation of at least one feature between a first image data and a second image data, the at least one feature including a distance between at least two elements detected at locations fixed in a background area and the first image data and the second image data including the head of the user and captured via an application executing on the computing device; and

predict, by a machine learning model accessible to the computing device, a fit of a first wearable device of a plurality of wearable devices on the head of the user based on the three-dimensional model of the head of the user;

wherein the predict by a machine learning model includes predicting a fit of the first wearable device according to at least one measurement extracted from the three-dimensional model and fit data indicating a fit of respective ones of the plurality of wearable devices on a plurality of users, the fit data based on

a plurality of different measurements of the plurality of users, wherein the at least one measurement extracted from the three-dimensional model corresponds to one or more of the plurality of different measurements.

19 . The system of claim 18 , wherein the instructions cause the at least one processor to:

generate the three-dimensional model of the head of the user;

extract at least one measurement from the three-dimensional model of the head of the user; and

select a wearable device, from a plurality of available wearable devices, based on the at least one measurement, the at least one measurement including at least one of:

a cranial measurement determined based on distance between two facial features detected in the second image data and the first image data; or

an ophthalmic measurement determined based on a distance between two optical features detected in the second image data and the first image data.

20 . The system of claim 18 , wherein the at least one feature includes a plurality of features, including:

at least one facial landmark defined by at least two facial features.