Eye movement tracking based on three-dimensional (3D) compressed sensing

Abstract

A three-dimensional (3D) compressive sensing based eye movement tracking system using a single photon avalanche diode (SPAD) sensor achieves high resolution depth measurements by using a low resolution SPAD sensor and actively coded illumination (e.g., two or three dimensional fringe patterns, random spots, random patterns, and / or overlay patterns) projected onto the surface of the eye. These patterns may be projected by a high speed illuminator, such as a digital micromirror device (DMD) or a micro-electro-mechanical system (MEMS) projector, varying continuously at the same rate as the SPAD sensor acquisition rate. The processor may employ compressed sensing techniques to obtain high resolution images and depth information from the acquired images of the varying patterns.

Description

Technical Field

[0001] This patent application generally relates to eye tracking in near-eye display devices, and more particularly to 3D compressive sensing-based eye tracking using single photon avalanche diode (SPAD) sensors. Background Art

[0002] With recent technological advancements, the popularity and dissemination of content creation and delivery have increased significantly in recent years. In particular, interactive content (such as virtual reality (VR) content, augmented reality (AR) content, mixed reality (MR) content), and content within and associated with a real environment and / or virtual environment (e.g., "virtual reality")) has become attractive to consumers.

[0003] To facilitate the delivery of such content and other related content, service providers have made efforts to provide various forms of wearable display systems. One such example can be a head-mounted display (HMD) device, such as wearable glasses, a wearable headset, or glasses. In some examples, the head-mounted display (HMD) device can project or direct light so that virtual objects can be displayed or images of real objects can be combined with virtual objects, as in virtual reality (VR) applications, augmented reality (AR) applications, or mixed reality (MR) applications. For example, in an AR system, a user can view an image of a virtual object (e.g., a computer-generated image (CGI)) and the surrounding environment simultaneously. The head-mounted display (HMD) device can also present interactive content, where the gaze of the user (wearer) can be used as an input to the interactive content. Summary of the Invention

[0004] According to a first aspect of the present invention, an eye tracking system includes: a projector configured to project a series of actively encoded illumination patterns onto an eye surface; and an SPAD-based sensor system configured to collect reflections of the series of actively encoded illumination patterns from the eye surface, wherein the projector and the SPAD-based sensor system are synchronized; compressive sensing is used to obtain a high-resolution image from the collected reflections; and three-dimensional (3D) features of the eye surface are obtained from the high-resolution image to determine the gaze of a user.

[0005] In some embodiments, the projector is operable to project the series of actively encoded illumination patterns onto the eye surface at a rate of up to 500 kilohertz (kHz).

[0006] In some embodiments, the SPAD-based sensor system includes at least one SPAD sensor, a linear array of SPAD sensors, or a two-dimensional array of SPAD sensors.

[0007] In some embodiments, the projector includes a laser source for providing collimated light and an ultrafast micro-electromechanical system (MEMS) reflector for projecting the series of actively encoded illumination patterns by directing the collimated light.

[0008] In some embodiments, the surface of the ultrafast MEMS reflector includes a plurality of diffractive optical elements or metasurfaces.

[0009] In some embodiments, the ultrafast MEMS reflector operates at a rate of up to 500 kHz.

[0010] In some embodiments, the projector includes a light source for providing light and a digital micromirror device (DMD) reflector for projecting the series of actively encoded illumination patterns by directing light from the light source.

[0011] In some embodiments, the light source includes a side-emitting laser diode, a vertical-cavity surface-emitting laser (VCSEL) diode, a superluminescent light-emitting diode (SLED), or a light-emitting diode (LED).

[0012] In some embodiments, the series of actively encoded illumination patterns includes two-dimensional encoded patterns, speckle patterns, random patterns, and combinations thereof.

[0013] In some embodiments, the plurality of actively encoded illumination patterns are selected based on at least one of pattern complexity, projector type, and sensor system type.

[0014] According to a second aspect of the present invention, a method includes: collecting, by a single-photon avalanche diode (SPAD)-based sensor system, reflections of a series of actively encoded illumination patterns projected onto the eye surface by a projector, wherein the projector and the SPAD-based sensor system are synchronized; using compressive sensing to obtain a high-resolution image from the collected reflections; obtaining three-dimensional (3D) features of the eye surface from the high-resolution image; and determining a user's gaze based on the obtained 3D features of the eye surface.

[0015] In some embodiments, the method further includes projecting a deterministic pattern or a random pattern onto the eye surface.

[0016] In some embodiments, the deterministic pattern includes an orthographic pattern or a two-dimensional encoded pattern and the random pattern includes a speckle pattern or a pseudo-random pattern.

[0017] In some embodiments, the method further includes selecting a plurality of actively encoded illumination patterns based on at least one of pattern complexity, projector type, and sensor system type.

[0018] In some embodiments, the method further includes: providing collimated light from a laser source to an ultrafast microelectromechanical systems (MEMS) reflector; and projecting the series of actively encoded illumination patterns by directing the collimated light from the ultrafast MEMS reflector onto the eye surface.

[0019] In some embodiments, the method further includes providing light from a light source to a digital micromirror device (DMD) reflector; and projecting the series of actively encoded illumination patterns by directing the light from the light source from the DMD reflector onto the eye surface.

[0020] According to a third aspect of the present invention, a non-transitory computer-readable storage medium has an executable file stored thereon, the executable file when executed instructing a processor to: project, by the projector, a series of actively encoded illumination patterns including orthographic deterministic patterns onto the eye surface at a rate of up to 500 kHz; collect, at a single-photon avalanche diode (SPAD)-based sensor system, reflections of the series of actively encoded illumination patterns from the eye surface, wherein the projector and the SPAD-based sensor system are synchronized; use compressive sensing to obtain a high-resolution image from the collected reflections; obtain three-dimensional (3D) features of the eye surface from the high-resolution image; and determine a user's gaze based on the obtained 3D features of the eye surface.

[0021] In some embodiments, when executed, the executable file instructs a processor to select a plurality of active encoding illumination patterns based on at least one of pattern complexity, projector type, and sensor system type.

[0022] In some embodiments, the series of active encoding illumination patterns further includes random patterns.

[0023] In some embodiments, the projector includes a microelectromechanical systems (MEMS) reflector or a digital micromirror device (DMD) reflector.

[0024] The method of the second aspect is, for example, a method of using the system of the first aspect, the system of the first aspect is, for example, a system for implementing the method of the second aspect, and the medium of the third aspect has an executable file for executing the method of the second aspect, and the preferred features and embodiments will be understood to apply correspondingly to these aspects. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Features of the present disclosure are shown by way of example and are not limited in the following drawings, in which like reference numerals indicate like elements. Those skilled in the art will readily recognize from the following that alternative examples of the structures and methods shown in the various drawings may be employed without departing from the principles described herein.

[0026] Figure 1 A block diagram of an artificial reality system environment including a near-eye display according to an example is shown.

[0027] Figure 2 A perspective view of a near-eye display in the form of a head-mounted display (HMD) device according to an example is shown.

[0028] Figure 3A and Figure 3B A perspective view and a top view of a near-eye display in the form of a pair of glasses according to an example are shown.

[0029] Figure 4 A schematic diagram of a structured light-based eye tracking system according to an example is shown.

[0030] Figure 5A and Figure 5B A configuration of a three-dimensional (3D) compressive sensing-based eye tracking system using single-photon avalanche diode (SPAD) sensors according to an example is shown.

[0031] Figure 6A and Figure 6B An example pattern that can be projected onto an eye surface for three-dimensional (3D) compressive sensing-based eye tracking according to an example is shown.

[0032] Figure 7A flowchart of a method for three-dimensional (3D) compressive sensing-based eye tracking using a single-photon avalanche diode (SPAD) sensor according to some examples is shown. DETAILED DESCRIPTION

[0033] For simplicity and illustrative purposes, the present application is described primarily with reference to examples of the present application. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. However, it will be apparent that the present application may be practiced without limitation to these specific details. In other instances, some methods and structures that are well understood by those of ordinary skill in the art are not described in detail so as not to unnecessarily obscure the present application. As used herein, the terms "a" and "an" are intended to denote at least one particular element, the term "comprising" means including but not limited to, the term "including" means including but not limited to, and the term "based on" means at least partially based on.

[0034] Tracking the position and orientation of the eyes and the gaze direction in a head-mounted display (HMD) device can provide a display and rendering structure that can significantly reduce the power requirements and computational requirements for rendering a 3D environment. In addition, gaze prediction and intent inference that support eye tracking can enable an intuitive and immersive user experience that adapts to the user's needs for their interaction with the virtual environment.

[0035] Eye tracking can be achieved by a variety of techniques. One technique is structured light projection, which projects a periodic pattern onto the eye and uses the reflected pattern to determine three-dimensional (3D) features. Another technique utilizes time-of-flight analysis of light projected onto the eye. These techniques and similar techniques involve the projection of light, e.g., projecting a laser onto the eye and collecting the reflection from the eye at close range.

[0036] Eye movements can be as high as 1000 degrees per second. Therefore, high-speed measurement is crucial for eye movement tracking applications. Thus, in order to use structured light sensing to achieve ultra-high-speed eye movement tracking of more than 1000 frames per second, both fast illumination and fast detection may be required. For augmented reality (AR) / virtual reality (VR) eye movement tracking applications, strict size and power requirements limit the use of traditional lighting devices (such as projectors based on digital light processing (DLP) and liquid crystal technology on silicon (LCOS)). It is also difficult to achieve such a high image acquisition speed using a camera (unless a dedicated camera is used). Specifically, the small form factor cameras used in near-eye display devices may not be able to achieve such an acquisition rate. Single-photon avalanche diode (SPAD) sensors are small form factor sensors that can achieve ultra-fast speeds. However, single-photon avalanche diode (SPAD) sensors have their own challenges in eye movement tracking applications, such as limited bandwidth, low resolution, and / or noise sensitivity, etc.

[0037] In some examples of the present disclosure, a three-dimensional (3D) compressed sensing-based eye movement tracking system using single-photon avalanche diode (SPAD) sensors is described. In order to achieve high-resolution depth measurement by using a low-resolution single-photon avalanche diode sensor, an active coded illumination such as a two-dimensional or three-dimensional stripe pattern, random dots, random patterns, and / or superimposed patterns can be projected onto the eye surface, and the reflection of the projected pattern can be collected by one or more single-photon avalanche diode sensors. A forward observation model and multiple measurements can be used to obtain a higher-resolution three-dimensional (3D) result. A high-speed illuminator such as a digital micromirror device (DMD) or a microelectromechanical system (MEMS) can be used to change the pattern at the same rate as one or more single-photon avalanche diode (SPAD) sensors. The processor can employ compressed sensing techniques to obtain high-resolution images and depth information from the images of the collected changing patterns. By quickly (synchronously) changing the projection pattern and the acquisition speed, high-resolution information can be collected at a faster rate.

[0038] Although some advantages and benefits of the present disclosure are obvious, other advantages and benefits may include: reducing computing resources and increasing the speed of eye movement tracking without increasing the complexity of the high-speed camera of the eye movement tracking system. In addition, the accuracy and / or power consumption efficiency of the eye movement tracking system can also be improved.

[0039] Figure 1FIG. 0 shows a block diagram of an artificial reality system environment 100 including a near-eye display according to an example. As used herein, a "near-eye display" may refer to a device (e.g., an optical device) that can be very close to a user's eyes. As used herein, "artificial reality" may refer to aspects such as "virtual reality" or an environment of real and virtual elements, and may include the use of technologies associated with virtual reality (VR), augmented reality (AR), and / or mixed reality (MR). As used herein, a "user" may refer to a user or wearer of a "near-eye display".

[0040] As Figure 1 shown, the artificial reality system environment 100 may include a near-eye display 120, an optional external imaging device 150, and an optional input / output interface 140, each of which may be coupled to a console 110. In some instances, the console 110 may be optional because the functions of the console 110 may be integrated into the near-eye display 120. In some examples, the near-eye display 120 may be a head-mounted display (HMD) that presents content to a user.

[0041] In some instances, for a near-eye display system, it may generally be desirable to enlarge the eye box, reduce display haze, improve image quality (e.g., resolution and contrast), reduce physical size, improve power efficiency, and increase or expand the field of view (FOV). As used herein, the "field of view" (FOV) may refer to the angular range of the image seen by the user, which is typically measured as the angle observed by one eye (for a monocular head-mounted display (HMD)) or two eyes (for a binocular head-mounted display (HMD)). Additionally, as used herein, the "eye box" may be a two-dimensional box that can be positioned in front of the user's eyes from which the image from an image source being displayed can be viewed.

[0042] In some examples, in a near-eye display system, light from the surrounding environment may pass through the "see-through" region (e.g., a transparent substrate) of the waveguide display to reach the user's eyes. For example, in a near-eye display system, light of the projected image may be coupled into the transparent substrate of the waveguide, propagate within the waveguide, and be coupled out or guided out of the waveguide at one or more locations to replicate the exit pupil and enlarge the eye box.

[0043] In some examples, the near-eye display 120 can include one or more rigid bodies that can be rigidly or non-rigidly coupled to each other. In some examples, the rigid coupling between the rigid bodies can cause the coupled rigid bodies to act as a single rigid entity, while in other examples, the non-rigid coupling between the rigid bodies can allow the rigid bodies to move relative to each other.

[0044] In some examples, the near-eye display 120 can be implemented in any suitable form factor, including a head-mounted display (HMD), a pair of glasses, or other similar wearable glasses or devices. Examples of the near-eye display 120 are further described below with reference to Figure 2 and FIG. 3. Additionally, in some examples, the functionality described herein can be used in a head-mounted display (HMD) or head-mounted device that can combine an image of the environment external to the near-eye display 120 with artificial reality content (e.g., computer-generated images). Thus, in some examples, the near-eye display 120 can utilize generated and / or superimposed digital content (e.g., images, videos, sounds, etc.) to enhance the image of the physical real-world environment external to the near-eye display 120 to present augmented reality to the user.

[0045] In some examples, the near-eye display 120 can include any number of display electronics 122, display optics 124, and an eye tracking unit 130. In some examples, the near-eye display 120 can also include one or more locators 126, one or more position sensors 128, and an inertial measurement unit (IMU) 132. In some examples, the near-eye display 120 can omit any of the eye tracking unit 130, one or more locators 126, one or more position sensors 128, and the inertial measurement unit (IMU) 132, or can include additional elements.

[0046] In some examples, the display electronics 122 can display an image to the user or facilitate the display of an image to the user based on data received, for example, from an optional console 110. In some examples, the display electronics 122 can include one or more display panels. In some examples, the display electronics 122 can include any number of pixels to emit light of primary colors (e.g., red, green, blue, white, or yellow). In some examples, the display electronics 122 can display a three-dimensional (3D) image, for example, using a stereoscopic effect produced by a two-dimensional panel to create a subjective perception of image depth.

[0047] In some examples, the near-eye display 120 can include a projector (not shown) that can form an image in an angular domain for direct viewing by a viewer's eye through the pupil. The projector can employ a controllable light source (e.g., a laser source) and a microelectromechanical systems (MEMS) beam scanner to create, for example, a light field from a collimated beam. In some examples, the same projector or a different projector can be used to project a stripe pattern onto the eye, which can be captured by a camera and analyzed (e.g., by the eye tracking unit 130) to determine the position, gaze, etc. of the eye (pupil).

[0048] In some examples, the display optics 124 can optically display image content (e.g., using optical waveguides and / or couplers), or magnify received image light from the display electronics 122, correct optical errors associated with the image light, and present the corrected image light to a user of the near-eye display 120. In some examples, the display optics 124 can include a single optical element or any number of combinations of various optical elements, as well as mechanical couplings for maintaining the relative spacing and orientation of the optical elements in the combination. In some examples, one or more of the optical elements in the display optics 124 can have an optical coating, such as an anti-reflection coating, a reflective coating, a filter coating, and / or a combination of different optical coatings.

[0049] In some examples, the display optics 124 can also be designed to correct one or more types of optical errors, such as two-dimensional optical errors, three-dimensional optical errors, or any combination thereof. Examples of two-dimensional errors can include barrel distortion, pincushion distortion, longitudinal chromatic aberration, and / or lateral chromatic aberration. Examples of three-dimensional errors can include spherical aberration, chromatic aberration of field curvature, and astigmatism.

[0050] In some examples, one or more locators 126 can be objects that are located at specific positions relative to each other and relative to a reference point on the near-eye display 120. In some examples, the optional console 110 can identify one or more locators 126 in an image captured by the optional external imaging device 150 to determine the position, orientation, or both of the artificial reality head-mounted device. One or more locators 126 can each be a light-emitting diode (LED), a corner cube reflector, a reflective marker, a type of light source that contrasts with the environment in which the near-eye display 120 operates, or any combination thereof.

[0051] In some examples, the external imaging device 150 may include one or more cameras, one or more video cameras, any other device capable of capturing an image including one or more locators 126, or any combination thereof. The optional external imaging device 150 may be configured to detect light emitted or reflected from one or more locators 126 within the field of view of the optional external imaging device 150.

[0052] In some examples, one or more position sensors 128 may generate one or more measurement signals in response to movement of the near-eye display 120. Examples of one or more position sensors 128 may include any number of accelerometers, gyroscopes, magnetometers, and / or other motion detection sensors or error correction sensors, or any combination thereof.

[0053] In some examples, an inertial measurement unit (IMU) 132 may be an electronic device that generates fast calibration data based on received measurement signals from one or more position sensors 128. One or more position sensors 128 may be located external to the inertial measurement unit (IMU) 132, within the inertial measurement unit (IMU) 132, or any combination thereof. The inertial measurement unit (IMU) 132 may generate fast calibration data indicative of an estimated position of the near-eye display 120 based on one or more measurement signals from one or more position sensors 128, which may be relative to an initial position of the near-eye display 120. For example, the inertial measurement unit (IMU) 132 may integrate received measurement signals from an accelerometer over time to estimate a velocity vector and integrate the velocity vector over time to determine an estimated position of a reference point on the near-eye display 120. Alternatively, the inertial measurement unit (IMU) 132 may provide sampled measurement signals to the optional console 110, which may determine the fast calibration data.

[0054] The eye tracking unit 130 may include one or more eye tracking systems. As used herein, "eye tracking" may refer to determining the position or relative position of an eye, which includes the orientation, position, and / or gaze of a user's eye. In some examples, an eye tracking system may include an imaging system that captures one or more images of an eye and may optionally include a light emitter that may generate light (e.g., a stripe pattern) directed at the eye such that light reflected by the eye may be captured by the imaging system (e.g., a camera). In other examples, the eye tracking unit 130 may capture reflected radio waves emitted by a micro radar unit. This eye-associated data may be used to determine or predict the position, orientation, movement, location, and / or gaze of the eye.

[0055] In some examples, the near-eye display 120 can use the orientation of the eyes to introduce depth cues (e.g., blurring images outside the user's primary line of sight), gather heuristics about user interactions in virtual reality (VR) media (e.g., the time spent on any particular subject, object, or frame based on the stimuli to which the user is exposed), some other functions that are partially based on the orientation of at least one of the user's eyes, or any combination thereof. In some examples, since the orientations of the user's two eyes can be determined, the eye tracking unit 130 can be able to determine where the user is looking or predict any user patterns, etc.

[0056] In some examples, the input / output interface 140 can be a device that allows the user to send action requests to the optional console 110. As used herein, an "action request" can be a request to perform a particular action. For example, an action request can be to start or end an application, or to perform a particular action within an application. The input / output interface 140 can include one or more input devices. Example input devices can include a keyboard, a mouse, a game controller, gloves, buttons, a touch screen, or any other suitable device for receiving an action request and transmitting the received action request to the optional console 110. In some examples, the action request received by the input / output interface 140 can be transmitted to the optional console 110, which can perform an action corresponding to the requested action.

[0057] In some examples, the optional console 110 can provide content to the near-eye display 120 for presentation to the user based on information received from one or more of the external imaging device 150, the near-eye display 120, and the input / output interface 140. For example, in Figure 1 the example shown, the optional console 110 can include an application repository 112, a head-mounted device tracking module 114, a virtual reality engine 116, and an eye tracking module 118. Some examples of the optional console 110 can include modules different from those described in connection with Figure 1 or additional modules in addition to those described in connection with Figure 1 The functions further described below can be distributed among the components of the optional console 110 in a manner different from the manner described herein.

[0058] In some examples, the optional console 110 can include a processor and a non-transitory computer-readable storage medium storing instructions executable by the processor. The processor can include multiple processing units that execute instructions in parallel. The non-transitory computer-readable storage medium can be any memory, such as a hard disk drive, removable memory, or solid-state drive (e.g., flash memory or dynamic random access memory (DRAM)). In some examples, in combination with Figure 1 the various modules of the optional console 110 described can be encoded as instructions in a non-transitory computer-readable storage medium, which when executed by the processor cause the processor to perform the functions further described below. It should be recognized that the optional console 110 may or may not be required, or the optional console 110 can be integrated with or separate from the near-eye display 120.

[0059] In some examples, the application repository 112 can store one or more applications for the optional console 110 to execute. An application can include a set of instructions that, when executed by the processor, generate content for presentation to the user. Examples of applications can include game applications, conferencing applications, video playback applications, or other suitable applications.

[0060] In some examples, the head-mounted device tracking module 114 can use slow calibration information from the external imaging device 150 to track the movement of the near-eye display 120. For example, the head-mounted device tracking module 114 can use the viewfinder locator to determine the position of the reference point of the near-eye display 120 based on the slow calibration information and the model of the near-eye display 120. Additionally, in some examples, the head-mounted device tracking module 114 can use fast calibration information, slow calibration information, or portions of any combination thereof to predict the future position of the near-eye display 120. In some examples, the head-mounted device tracking module 114 can provide the estimated or predicted future position of the near-eye display 120 to the virtual reality engine 116.

[0061] In some examples, the virtual reality engine 116 can execute an application within the artificial reality system environment 100 and receive position information of the near-eye display 120, acceleration information of the near-eye display 120, velocity information of the near-eye display 120, the predicted future position of the near-eye display 120, or any combination thereof from the head-mounted device tracking module 114. In some examples, the virtual reality engine 116 can also receive the estimated eye position and orientation information from the eye tracking module 118. The virtual reality engine 116 can determine the content to provide to the near-eye display 120 for display to the user based on the received information.

[0062] In some examples, the eye tracking module 118, which can be implemented as a processor, can receive eye tracking data from the eye tracking unit 130 and determine the position of the user's eyes based on the eye tracking data. In some examples, the eye position can include the orientation, position, or both of the eyes relative to the near-eye display 120 or any element of the near-eye display. Thus, in these examples, since the axis of rotation of the eyes changes according to the position of the eyes in their orbits, determining the position of the eyes in their orbits can allow the eye tracking module 118 to more accurately determine the orientation of the eyes.

[0063] In some examples, the position of the projector of the display system can be adjusted to implement any number of design modifications. For example, in some instances, the projector can be located in front of the viewer's eyes (i.e., "front-mounted" placement). In a front-mounted placement, in some examples, the projector of the display system can be located at a position away from the user's eyes (i.e., the "world side"). In some examples, a head-mounted display (HMD) device can utilize a front-mounted placement to project light toward one or both of the user's eyes to project an image.

[0064] Figure 2 A perspective view of a near-eye display in the form of a head-mounted display (HMD) device 200 according to one example is shown. In some examples, the head-mounted display (HMD) device 200 can be part of the following systems: a virtual reality (VR) system, an augmented reality (AR) system, a mixed reality (MR) system, another system using a display or a wearable device, or any combination thereof. In some examples, the head-mounted display (HMD) device 200 can include a display 210, a body 220, and a headband 230. Figure 2 The bottom side 223, front side 225, and left side 227 of the body 220 are shown in the perspective view. In some examples, the headband 230 can have an adjustable or extendable length. In particular, in some examples, there can be sufficient space between the body 220 of the head-mounted display (HMD) device 200 and the headband 230 to allow the user to secure the head-mounted display (HMD) device 200 to the user's head. For example, the length of the headband 230 can be adjustable to accommodate a range of user head sizes. In some examples, the head-mounted display (HMD) device 200 can include additional components, fewer components, and / or different components.

[0065] In some examples, a head-mounted display (HMD) device 200 can present media or other digital content to a user, the media or other digital content including a virtual view with computer-generated elements and / or an augmented view of a physical real-world environment. Examples of media or digital content presented by the head-mounted display (HMD) device 200 can include images (e.g., two-dimensional (2D) images or three-dimensional (3D) images), videos (e.g., 2D videos or 3D videos), audio, or any combination thereof. In some examples, images and videos can be presented to each eye of the user via one or more display components ( Figure 2 not shown in), the one or more display components being enclosed within a body 220 of the head-mounted display (HMD) device 200.

[0066] In some examples, the head-mounted display (HMD) device 200 can include various sensors (not shown), such as depth sensors, motion sensors, position sensors, and / or eye-tracking sensors. Some of these sensors can use any number of structured light patterns or unstructured light patterns for sensing purposes. In some examples, the head-mounted display (HMD) device 200 can include: an input / output interface 140 for communicating with a console 110 as described with reference to Figure 1 . In some examples, the head-mounted display (HMD) device 200 can include a virtual reality engine (not shown), and the virtual display engine is similar to the virtual reality engine 116 described with reference to Figure 1 , the virtual reality engine being capable of executing applications within the head-mounted display (HMD) device 200 and receiving depth information, position information, acceleration information, velocity information, predicted future positions, or any combination thereof of the head-mounted display (HMD) device 200 from various sensors.

[0067] In some examples, the information received by the virtual reality engine 116 can be used to generate signals (e.g., display instructions) to one or more display components. In some examples, the head-mounted display (HMD) device 200 can include a plurality of locators (not shown), and the plurality of locators is similar to the locators 126 described in Figure 1 , the plurality of locators being located at fixed positions on the body 220 of the head-mounted display (HMD) device 200 relative to each other and relative to a reference point. Each of these locators can emit light detectable by an external imaging device. This can be useful for head tracking or other motion / orientation purposes. It should be recognized that other elements or components can be used in addition to or in place of such locators.

[0068] It should be recognized that, in some examples, a projector installed in a display system can be placed near and / or close to a user's eyes (i.e., "eye side"). In some examples, and as discussed herein, a projector of a display system shaped like glasses can be installed or positioned in the temple arms of the glasses (i.e., the top far corner on the lens side). It should be recognized that, in some instances, using a rear-mounted projector placement can help reduce the size or volume of any required housing for the display system, which can also result in a significant improvement in the user experience of the user.

[0069] In some examples, the projector can provide structured light (a stripe pattern) onto the eye, which can be captured by the eye tracking sensor 212. The eye tracking sensor 212 or a communicatively coupled processor (e.g., Figure 1 the eye tracking module 118 in

[0070] Figure 3A is a perspective view 300A of a near-eye display 300 in the form of a pair of glasses (or other similar glasses) according to an example. In some examples, the near-eye display 300 can be Figure 1 a specific example of the near-eye display 120 in

[0071] In some examples, the near-eye display 300 can include a frame 305 and a display 310. In some examples, the display 310 can be configured to present media or other content to the user. In some examples, similar to the components described with reference to Figure 1 and Figure 2 the display 310 can include display electronics and / or display optics. For example, as referred to above with reference to Figure 1As described for the near-eye display 120 in [reference], the display 310 may include a liquid crystal display (LCD) panel, a light emitting diode (LED) display panel, or an optical display panel (e.g., a waveguide display assembly). In some examples, the display 310 may also include any number of optical components such as waveguides, gratings, lenses, mirrors, etc. In other examples, the display 310 may include a projector; or alternatively, the near-eye display 300 may include a projector instead of the display 310.

[0072] In some examples, the near-eye display 300 may also include various sensors 350a, 350b, 350c, 350d, and 350e on or within the frame 305. In some examples, as shown, the various sensors 350a to 350e may include any number of depth sensors, motion sensors, position sensors, inertial sensors, and / or ambient light sensors. In some examples, the various sensors 350a to 350e may include any number of image sensors configured to generate image data representing different fields of view in one or more different directions. In some examples, the various sensors 350a to 350e may be used as input devices to control or affect the display content of the near-eye display and / or to provide an interactive virtual reality (VR) experience, augmented reality (AR) experience, and / or mixed reality (MR) experience to the user of the near-eye display 300. In some examples, the various sensors 350a to 350e may also be used for stereoscopic imaging or other similar applications.

[0073] In some examples, the near-eye display 300 may also include one or more illuminators 330 for projecting light into the physical environment. The projected light may be associated with different frequency bands (e.g., visible light, infrared light, ultraviolet light, etc.) and may be used for various purposes. In some examples, one or more illuminators 330 may be used as locators, such as one or more locators 126 described above with reference to Figure 1 and Figure 2 the one or more locators described.

[0074] In some examples, the near-eye display 300 may also include a camera 340 or other image acquisition unit. For example, the camera 340 may acquire an image of the physical environment in the field of view. In some instances, it may be, for example, by a virtual reality engine (e.g., Figure 1The virtual reality engine 116) in processes the captured images to add virtual objects to the captured images or modify physical objects in the captured images, and for augmented reality (AR) applications and / or mixed reality (MR) applications, the processed images can be displayed to the user by the display 310. The near-eye display 300 may also include an eye tracking sensor 312.

[0075] Figure 3B FIG. 300B is a top view of a near-eye display 300 in the form of a pair of glasses (or other similar eyewear) according to one example. In some examples, the near-eye display 300 may include a frame 305 having the shape elements of a pair of glasses. For each eye, the frame 305 supports: a stripe projector, such as any stripe projector variant contemplated herein; a display 310 for presenting content to the appropriate eye region 366; an eye tracking sensor 312; and one or more illuminators 330. The illuminator 330 may be used to illuminate the appropriate eye region 366 and to provide flash illumination to the eye. The stripe projector 314 may provide a periodic stripe pattern onto the user's eye. The display 310 may include a pupil replication waveguide to receive a fan of light beams and provide multiple laterally offset parallel copies of each beam in the fan of light beams, thereby extending the projected image onto the appropriate eye region 366.

[0076] In some examples, the pupil replication waveguide may be transparent or translucent to enable the user to view the external world and the image projected into each eye and superimposed on the external world view. The image projected into each eye may include objects disposed with simulated parallax so as to appear immersed in the real-world view.

[0077] The eye tracking sensor 312 may be used to determine the position and / or orientation of the user's binocular eyes. Once the position and orientation of the user's eyes are known, the vergence distance and direction of the gaze can be determined. In some examples, the eye tracking sensor 312 may be a single photon avalanche diode (SPAD) sensor. For better fidelity of the user's immersion in the displayed augmented reality scene and / or to provide specific functions for interacting with the augmented reality, the images displayed by the display 310 can be dynamically adjusted to account for the user's gaze. In operation, the illuminator 330 may illuminate the eye at the corresponding appropriate eye region 366 to enable the eye tracking camera to acquire an image of the eye and to provide a reference reflection. The reflection (also referred to as "flash") may serve as a reference point in the captured eye image to facilitate determination of the eye gaze direction by determining the position of the eye pupil image relative to the flash. To avoid distracting the user with the illumination light, the illumination light may be made invisible to the user. For example, infrared light may be used to illuminate the appropriate eye region 366.

[0078] In some examples, an image processing and eye position / orientation determination function may be provided with control signals by a central controller (not shown) 310 of the near-eye display 300 to generate an image to be displayed to the user.

[0079] Figure 4 A schematic diagram 400 of a structured light-based eye tracking system according to an example is shown. The schematic diagram 400 shows a light source D projecting a projector stripe 402(C) with phase lines 404 and projector pixels 406 onto an object 408(B) with phase lines 412, where the projector pixels 406 are projected as object points 410. The reflected stripe pattern is captured by a camera (sensor) E as a camera image 414(A), where the object points 410 are captured as camera pixels 416.

[0080] In some examples, at least one single-photon avalanche diode (SPAD) sensor may be used for fast image detection. In low-bit single-photon avalanche diode (SPAD)-based imaging (e.g., 1-bit to 3-bit), ultra-high speeds of up to several tens of thousands of frames per second can be achieved based on the single-photon avalanche diode (SPAD) photon counting ability. However, due to the relatively low photon detection efficiency (PDE) and false triggering caused by ambient noise, averaging may be required, and thus the effective 3D imaging frame rate may be low. To obtain a high-resolution image from a low-resolution image captured by a single-photon avalanche diode (SPAD)-based sensor, compressive sensing techniques may be employed to improve the image quality.

[0081] A single-photon avalanche diode (SPAD) sensor can reduce the integration time by increasing the gain, thereby enabling ultra-fast measurements. Due to the high-gain characteristics, a single-photon avalanche diode (SPAD) sensor is vulnerable to ambient light, resulting in noise in the measurements. In some examples, a narrowband single-photon avalanche diode (SPAD) sensor may be used to address the noise challenges in the feature matching process and achieve more accurate measurements.

[0082] Feature matching refers to finding corresponding features from two similar images based on a search distance algorithm. One of the two images can be considered the source image, and the other can be considered the target image, and feature matching techniques can be used to find or derive multiple attributes and transfer these attributes from the source image to the target image. The feature matching process can analyze the topological structures of the source and target images, detect feature patterns, match the patterns, and match the features within the found patterns. The accuracy of feature matching may depend on image similarity, complexity, and quality. Thus, reduction of noise caused by ambient light in a single photon avalanche diode (SPAD) sensor can provide increased accuracy in detecting three-dimensional (3D) features of an eye (specifically the pupil), and then the three-dimensional (3D) features of the eye can be used to determine the user's gaze.

[0083] Figure 5A and Figure 5B illustrates a configuration of an eye tracking system based on three-dimensional (3D) compressive sensing using a single photon avalanche diode (SPAD) sensor according to an example. Figure 5A Schematic diagram 500A therein illustrates a microelectromechanical system (MEMS)-based projector 504 projecting active coded illumination onto the surface of an eye 502, and a reflection of the projected pattern being acquired by a low-resolution single photon avalanche diode (SPAD) sensor 506. The microelectromechanical system (MEMS)-based projector 504 and the single photon avalanche diode (SPAD) sensor 506 can be communicatively coupled to a processor 510, which can synchronize the microelectromechanical system (MEMS)-based projector 504 and the single photon avalanche diode (SPAD) sensor 506 in time and process the acquired images using compressive sensing techniques.

[0084] A microelectromechanical system (MEMS) can be composed of components sized between 1 micrometer and 100 micrometers and includes a central unit for processing data (e.g., an integrated circuit chip such as a microprocessor) and several components for interacting with the surrounding environment (e.g., micromirrors or reflectors). Movement can be achieved using electrostatic charge or magnetic moment. In the microelectromechanical system (MEMS) projector 504, light provided by a collimated light source such as a side-emitting laser diode, a vertical-cavity surface-emitting laser (VCSEL) diode, or a superluminescent light-emitting diode (SLED) can be reflected by at least one microelectromechanical system (MEMS) reflector onto the eye. The microelectromechanical system (MEMS) reflector can be moved or scanned in one or more directions to project an active coded illumination pattern onto the eye surface. The microelectromechanical system (MEMS) reflector can operate, for example, between 10 kHz and 100 kHz, direct the laser to the far field, and generate a desired pattern on the eye surface by controlling the laser intensity and timing.

[0085] In some examples, instead of generating vertical lines, a two-dimensional pattern can be generated by a diffractive optical element (DOE) or a metasurface on the surface of a microelectromechanical system (MEMS) reflector. This method may be effective when only a limited number of longitudinal resolutions are required for eye tracking. A focused beam in certain directions may also help improve the signal-to-noise ratio.

[0086] A single-photon avalanche diode (SPAD) photodetector is implemented using CMOS technology. The p-n junction of the single-photon avalanche diode (SPAD) photodetector is reverse-biased above its breakdown voltage, such that a single photon incident on the active device region can generate an electron-hole pair, thereby triggering a secondary carrier avalanche. The avalanche occurrence time can be on the order of picoseconds, so the associated voltage change can be used to precisely measure the arrival time of the photon. Although a single-photon avalanche diode (SPAD) sensor can acquire images at a very high rate, it has low resolution. For example, a single-photon avalanche diode (SPAD) sensor may have a resolution of 32×32 pixels. Compared with traditional cameras that may have bit rates of 8, 12, 16, or higher, a typical single-photon avalanche diode (SPAD) sensor can have a bit rate between 1 and 4 bits.

[0087] Figure 5B The schematic diagram 500B in shows a configuration similar to the configuration in schematic diagram 500A, but instead of a microelectromechanical system (MEMS)-based projector 504 for projecting active coded illumination onto the eye 502, a digital micromirror device (DMD) projector 514 is used for projecting the active coded illumination onto the eye.

[0088] A digital micromirror device (DMD) uses a large number of tiny movable mirrors to form images. A digital micromirror device (DMD) chip can have hundreds of thousands of micromirrors arranged in a rectangular array on its surface, and the hundreds of thousands of micromirrors correspond to the pixels of the image to be projected. In the on or off state, the mirrors can rotate ±10 - 12 degrees. In the on state, light from the light source may be reflected by the mirrors, making the pixels appear bright. In the off state, the light may be directed elsewhere, making the pixels appear dark. The on and off states of the mirrors can be switched very quickly (e.g., 1 to 5 microseconds). The light source of the digital micromirror device (DMD) can be a light-emitting diode (LED), a side-emitting laser diode, a vertical-cavity surface-emitting laser (VCSEL) diode, a superluminescent light-emitting diode (SLED), or a similar source.

[0089] The mirrors in a digital micromirror device (DMD) can be made of aluminum and have a diameter of about 15 micrometers in some embodiments. Each mirror can be mounted on a yoke which can in turn be connected to two support posts by compliant torsion hinges. Two pairs of electrodes can control the position of the mirror by electrostatic attraction. Each pair of electrodes can have one electrode on each side of the hinge, where one of the two pairs is positioned to act on the yoke and the other pair acts directly on the mirror. In some embodiments, the state of the mirror can be loaded into a memory (e.g., static random access memory, “SRAM”) which can also be connected to the electrodes. Once all the SRAM cells (for all the mirrors) are loaded, the bias voltage can be removed, allowing the charge from the SRAM cells to prevail and thus move the mirror. When the bias is restored, the mirror can remain in place again and the next movement can be loaded into the corresponding storage cell.

[0090] As mentioned herein, a single photon avalanche diode (SPAD) sensor 506 can achieve an image acquisition rate of up to 500 kHz, but operates at more than 1 or 4 bits, acquiring 32×32 pixels at a time. A processor 510 can time synchronize (508) the single photon avalanche diode (SPAD) sensor 506 and the digital micromirror device (DMD) projector 514 (or the microelectromechanical system (MEMS) projector 504 in schematic 500A) to ensure acquisition of a series of active coded illumination patterns for fast projection. Then, the processor 510 can use compressive sensing techniques to obtain a high-resolution image (e.g., 256×256) from the images acquired by the single photon avalanche diode (SPAD) sensor 506.

[0091] In some examples, multiple single photon avalanche diode (SPAD) sensors can be used to acquire the reflection of a pattern from the eye surface. In other examples, a linear array of single photon avalanche diode (SPAD) sensors combined with an active microelectromechanical system (MEMS) shutter array or a two-dimensional single photon avalanche diode (SPAD) array can also be used to acquire the reflection.

[0092] Figure 6A and Figure 6B illustrates example patterns that can be projected onto the eye surface for three-dimensional (3D) compressive sensing-based eye tracking. Figure 6A in schematic 600A and Figure 6B in schematic 600B include example active coded illumination patterns 602, 604, 606, and 608.

[0093] In some examples, the active encoding illumination patterns can include various arrangements of the following patterns: a standard stripe pattern (vertical or horizontal lines) superimposed on another pattern (e.g., pattern 602), a two-dimensional encoding pattern, a speckle pattern, a random pattern, and combinations thereof. Different patterns (e.g., pattern 606) can be sequentially projected at a rapid rate (e.g., up to 500 kHz) and acquired synchronously by one or more single-photon avalanche diode (SPAD) sensors. Some pattern sequences can be variants of a similar pattern type (e.g., pattern 606 or pattern 608).

[0094] Other pattern sequences can include random patterns or pseudo-random patterns. Each different pattern can provide different information, and thus more information about the eye surface can be acquired by using a series of varying patterns. In some examples, deterministic basis patterns (e.g., Hadamard pattern sequences) can be used. Such basis patterns can form a complete orthonormal set and allow the acquisition of spatial information of an object (eye surface) image in the transform domain. When the image is fully sampled in the transform domain, it can be reconstructed losslessly by the corresponding inverse transform. Multiple patterns can be selected based on pattern complexity and / or system configuration (projector, sensor type).

[0095] Compressed sensing is used to efficiently acquire and reconstruct an image (or other signal) by finding the solution of an underdetermined linear system based on the principle that, through optimization, the sparsity of the signal can be exploited to recover the signal from fewer samples than required by the Nyquist-Shannon sampling theorem. For reconstruction using compressed sensing, sparsity and incoherence are required conditions, both of which exist in the eye movement tracking pattern projection and acquisition system. The image transformation (e.g., wavelet transform) of the acquired eye surface satisfies the sparsity condition (i.e., the coefficients are generally small). Using a sequence of varying (e.g., orthogonal) patterns can satisfy the incoherence requirement.

[0096] Compressed sensing can start with a weighted linear combination of samples from a domain different from the domain in which the signal is sparse. To transform the image back to the desired domain, an underdetermined matrix equation can be solved because the number of compressive measurements is less than the number of pixels in the image. However, since the initial signal is sparse, the matrix equation can be solved as an underdetermined system of linear equations. Some example compressed sensing techniques may include, but are not limited to, convex optimization and adaptive gradient-based techniques, which can solve non-convex combinatorial optimization problems. Other example techniques may include greedy algorithms (a set of algorithms with lower computational complexity for obtaining the sparsest solution of the set (system)); and threshold-based algorithms (iterative hard thresholding and soft thresholding, automatic thresholding, etc.), which are based on an adaptive threshold applied in multiple iterations.

[0097] Figure 7 FIG. 700 is a flow chart of a method for three-dimensional (3D) compressive sensing based eye tracking using a single photon avalanche diode (SPAD) sensor, according to some examples. Since there may be various ways to perform the methods described herein, method 700 is provided by way of example. Although method 700 is primarily described as being performed by Figure 5A , Figure 5B components in, method 700 may be performed by one or more processing components of another system or combination of systems, or otherwise. Figure 7 Each block shown in may also represent one or more processes, one or more methods, or one or more subroutines, and one or more of these blocks (e.g., the selection process) may include machine-readable instructions stored on a non-transitory computer-readable medium and executed by a processor or other type of processing circuitry to perform one or more operations described herein.

[0098] At block 702, an active coded illumination pattern, such as a two-dimensional coded pattern, a speckle pattern, a random pattern, and combinations thereof, may be projected onto the eye surface at a high rate (e.g., up to 500 kHz). The projector may include a laser source and an ultrafast scanning microelectromechanical system (MEMS)-based projector or a digital micromirror device (DMD) projector with a suitable light source.

[0099] At block 704, one or more 2D single photon avalanche diode (SPAD) sensors or a single photon avalanche diode (SPAD) array detector may be used to collect the reflection of the projected pattern from the eye surface. One or more single photon avalanche diode (SPAD) sensors may be low-resolution sensors, e.g., operating in 1-bit or 4-bit mode, and time synchronized with the projector such that one or more sensors collect each projected pattern.

[0100] At block 706, compressive sensing techniques may be applied to the image of the collected reflection pattern. The image transformation (e.g., wavelet transform) of the collected eye surface satisfies the sparsity condition, and the sequence of varying (e.g., orthogonal) patterns may satisfy the incoherence requirement of compressive sensing. The high-resolution image and depth information may be retrieved from the collected image by compressive sensing.

[0101] At block 708, 3D features of the eye surface can be retrieved from the high-resolution information. These features can be retrieved by feature matching, which finds corresponding features from two similar images based on a search distance algorithm. One of the two images can be considered the source image, and the other can be considered the target image, and feature matching techniques can be used to find or derive multiple attributes and transfer these attributes from the source image to the target image. The feature matching process can analyze the topologies of the source and target images, detect feature patterns, match the patterns, and match the features within the discovered patterns. Then, the user's gaze can be inferred based on the retrieved features.

[0102] According to an example, a method of forming an eye tracking system based on three-dimensional (3D) compressive sensing using a single-photon avalanche diode (SPAD) sensor is described herein. A system for forming an eye tracking system is also described herein. An executable file can be stored on a non-transitory computer-readable storage medium, and when executed, the executable file instructs a processor to perform the methods described herein.

[0103] In the foregoing description, various examples have been described, including devices, systems, and methods, etc. For purposes of explanation, specific details have been set forth to provide a thorough understanding of the various examples of the present disclosure. However, it will be apparent that the various examples can be practiced without these specific details. For example, devices, systems, structures, components, methods, and other elements can be shown as components in block diagram form so as not to obscure the examples in unnecessary detail. In other instances, well-known devices, processes, systems, structures, and techniques may be shown without unnecessary detail so as not to obscure the examples.

[0104] The drawings and the specification are not intended to be restrictive. The terms and expressions employed in this disclosure are used as descriptive terms and not of a restrictive nature, and in using such terms and expressions, there is no intention to exclude any equivalents of the features shown and described or portions thereof. The word "example" is used herein to mean "serving as an example, instance, or illustration". Any embodiment or design described herein as an "example" is not necessarily to be construed as preferred or advantageous over other embodiments or designs.

[0105] Although the methods and systems described herein may be primarily directed to digital content (e.g., video or interactive media), it should be recognized that the methods and systems described herein can also be used for other types of content or scenarios. Other applications or uses of the methods and systems described herein can also include social networks, marketing, content-based recommendation engines, and / or other types of knowledge systems or data-driven systems.

Claims

1. An eye tracking system, the eye tracking system comprising: a projector configured to project a series of actively encoded illumination patterns onto the eye surface; and a single photon avalanche diode (SPAD)-based sensor system configured to collect reflections of the series of actively encoded illumination patterns from the eye surface, wherein the projector is synchronized with the SPAD-based sensor system; compressed sensing is used to obtain a high-resolution image from the collected reflections; and three-dimensional (3D) features of the eye surface are obtained from the high-resolution image to determine the user's gaze.

2. The eye tracking system according to claim 1, wherein the projector projects the series of actively encoded illumination patterns onto the eye surface at a rate of up to 500 kHz.

3. The eye tracking system according to claim 1 or claim 2, wherein the SPAD-based sensor system includes at least one SPAD sensor, a linear SPAD sensor array, or a two-dimensional SPAD sensor array.

4. The eye tracking system according to any one of the preceding claims, wherein the projector comprises: a laser source configured to provide collimated light; and an ultrafast microelectromechanical system (MEMS) reflector configured to project the series of actively encoded illumination patterns by directing the collimated light; and optionally, wherein the surface of the ultrafast MEMS reflector includes a plurality of diffractive optical elements or metasurfaces; and / or optionally, wherein the ultrafast MEMS reflector operates at a rate of up to 500 kHz.

5. The eye tracking system according to any one of the preceding claims, wherein the projector comprises: a light source configured to provide light; and a digital micromirror device (DMD) reflector configured to project the series of actively encoded illumination patterns by directing light from the light source; and optionally, wherein the light source includes a side-emitting laser diode, a vertical-cavity surface-emitting laser (VCSEL) diode, a superluminescent light-emitting diode (SLED), or a light-emitting diode (LED).

6. The eye tracking system according to any one of the preceding claims, wherein the series of actively encoded illumination patterns includes two-dimensional encoded patterns, speckle patterns, random patterns, and combinations thereof; and optionally, wherein a plurality of actively encoded illumination patterns are selected based on at least one of pattern complexity, projector type, and sensor system type.

7. A method, the method comprising: collecting, by a single photon avalanche diode (SPAD)-based sensor system, reflections of a series of actively encoded illumination patterns projected onto an eye surface by a projector, wherein the projector and the SPAD-based sensor system are synchronized; applying compressed sensing to obtain a high-resolution image from the collected reflections; obtaining three-dimensional (3D) features of the eye surface from the high-resolution image; and Determine the user's gaze based on the obtained 3D features of the eye surface.

8. The method according to claim 7, the method further comprises: Project a deterministic pattern or a random pattern onto the eye surface, wherein optionally, the deterministic pattern comprises an orthogonal pattern or a two-dimensional coded pattern, and the random pattern comprises a speckle pattern or a pseudo-random pattern.

9. The method according to claim 7 or 8, the method also comprises: Select a plurality of active coded illumination patterns based on at least one of pattern complexity, projector type, and sensor system type.

10. The method according to any one of claims 7 to 9, the method also comprises: Provide collimated light from a laser source to an ultrafast microelectromechanical system (MEMS) reflector; and, Project the series of active coded illumination patterns by directing the collimated light from the ultrafast MEMS reflector onto the eye surface.

11. The method according to any one of claims 7 to 10, the method also comprises: Provide light from a light source to a digital micromirror device (DMD) reflector; and Project the series of active coded illumination patterns by directing the light from the light source from the DMD reflector onto the eye surface.

12. A non-transitory computer-readable storage medium having an executable file stored thereon, the executable file when executed instructs a processor to: Project a series of active coded illumination patterns including orthogonal deterministic patterns onto the eye surface at a rate of up to 500 kHz through a projector; Collect reflections of the series of active coded illumination patterns from the eye surface at a single-photon avalanche diode (SPAD)-based sensor system, wherein, the projector and the SPAD-based sensor system are synchronized; Obtain a high-resolution image from the collected reflections using compressive sensing; Obtain three-dimensional (3D) features of the eye surface from the high-resolution image; and Determine the user's gaze based on the obtained 3D features of the eye surface.

13. The non-transitory computer-readable storage medium according to claim 12, wherein, the executable file when executed instructs the processor to: select a plurality of active coded illumination patterns based on at least one of pattern complexity, projector type, and sensor system type.

14. The non-transitory computer-readable storage medium according to claim 12 or 13, wherein, the series of active coded illumination patterns further comprises a random pattern.

15. The non-transitory computer-readable storage medium according to any one of claims 12 to 14, wherein, the projector comprises a microelectromechanical system (MEMS) reflector or a digital micromirror device (DMD) reflector.

Images