On Thursday, he U.S. Patent and Trademark Office officially published a patent application from Apple that reveals how privacy extends to a users eye. Apple describes a personalized eye model that contains privacy-and-security-sensitive information, and how it is stored as encrypted data when the user is not logged in to the device. Apple describes this as the N-Dimension Model

Synthetic Gaze Enrollment

An XR system may include an HMD which may include one or more cameras that may be used to capture still images or video frames of the user's environment. The HMD may include lenses positioned in front of the eyes through which the wearer can view the environment. In XR systems, virtual content may be displayed on or projected onto these lenses to make the virtual content visible to the wearer while still being able to view the real environment through the lenses.

In at least some systems, the HMD may include gaze tracking technology. In an example gaze tracking system, one or more infrared (IR) light sources emit IR light towards a user's eye. A portion of the IR light is reflected off the eye and captured by an eye tracking camera. Images captured by the eye tracking camera may be input to a glint and pupil detection process, for example implemented by one or more processors of a controller of the HMD. Results of the process are passed to a gaze estimation process, for example implemented by one or more processors of the controller, to estimate the user's current point of gaze. This method of gaze tracking may be referred to as PCCR (Pupil Center Corneal Reflection) tracking.

In such systems, during an initial calibration or enrollment process, a multidimensional personalized model of the user's eye may be generated from one or more images of the eye captured as described above. This personalized eye model may then be used in various algorithms, for example in the gaze estimation process, during use of the device. The personalized eye model may include information such as a cornea surface model, iris and pupil model, eye center, entrance pupil, pupillary or optical axis (a vector which passes through the geometric eye center and the entrance pupil), and a kappa angle between the optical axis and the visual axis.

However, a personalized eye model may be privacy- and security-sensitive information, and thus may be stored as encrypted data when the user is not logged in to the device. Thus, after a cold boot of the device, gaze enrollment data including the personalized eye model is not accessible due to the security and privacy concerns until the user logs in using a passcode or other secure login method, after which the secured data can be encrypted. 

However, entering a passcode to log in may be performed via a gaze-based interface. Thus, to calibrate the device for gaze interaction before login, since the personalized eye model is not available, a privacy-insensitive gaze calibration model is needed to enable or improve gaze-based passcode interaction for user login. This gaze calibration model, which may be referred to as a screen-space model, may be estimated in a gaze enrollment process using a synthetic gaze. 

The gaze correction function does not contain security- and privacy-sensitive information of the user. Further, since the method collapses a hyperdimensional space (the N-dimensional personalized eye model) into two dimensions (the gaze correction function), the personalized eye model cannot be recovered from the gaze correction function, and thus the gaze correction function can be stored unencrypted and available for use during a cold boot of the device prior to login.

The N-Dimension Model

Apple's patent FIG. 1 above graphically illustrates an N-dimensional model #100 of an eye, according to some embodiments. Physical components of an eye may include a sclera #102, cornea #104, iris #106, and pupil #108. In some embodiments, during an initial calibration or enrollment process, an N-dimensional model of the user's eye may be generated from one or more images of the eye.

In an example method, one or more infrared (IR) light sources emit IR light towards a user's eye. A portion of the IR light is reflected off the eye and captured by an eye tracking camera. Two or more images captured by the eye tracking camera may be input to an eye model generating process, for example implemented by one or more processors of a controller of the HMD. The process may determine the shapes and relationships of the eye's components based at least in part on positions of the glints (reflections of the point light sources) in the two or more captured images. This information may then be used to generate the personalized eye model.

The personalized eye model may include information such as a cornea surface model, iris and pupil model, eye center #112, entrance pupil #110, pupillary or optical axis #120 (a vector which passes through the eye center and the entrance pupil), and a kappa angle between the optical axis and the visual axis #122 of the eye. This personalized eye model may then be used in various algorithms, for example in the gaze estimation process, during use of the device. The personalized eye model may then then used in various algorithms, for example in the gaze estimation process, during use of the device.

For full details, review Apple's patent application  20240104967.