Correcting pupil center shift to calculate gaze

Abstract

Systems and methods for computing a gaze are provided. The systems and methods access an uncorrected gaze vector computed based on a center point of a user's pupil. The systems and methods process an image of the pupil through a machine learning model to predict an estimated error in the gaze vector, the machine learning model trained to establish, for a plurality of pupil parameters, a relationship between a plurality of ground truth gaze vectors and the uncorrected gaze vector, the pupil parameters including a diameter, a gaze angle, or a gaze eccentricity. The systems and methods generate a corrected gaze vector by applying the estimated error in the gaze vector predicted by the machine learning model to the uncorrected gaze vector that has been computed based on the center point of the user's pupil.

Description

[0001] Priority Statement

[0002] This application claims the benefit of priority to U.S. Provisional Application No. 63 / 595,588, filed November 2, 2023, and U.S. Patent Application No. 18 / 920,368, filed October 18, 2024, each of which is incorporated herein by reference in its entirety. Background Technology

[0003] Some electronically enabled glasses devices (such as so-called smart glasses) allow users to interact with virtual content (e.g., augmented reality (AR) objects) while engaging in certain activities. Users wear these glasses and can view the real-world environment through them while simultaneously interacting with virtual content displayed on the glasses. Certain electronically enabled glasses devices (and other AR devices) enable users to interact with virtual (or real-world) content based on tracking their eye gaze (e.g., tracking / determining where the user is looking in the environment presented to them). Attached Figure Description

[0004] The accompanying drawings are merely examples of this disclosure and should not be considered as limiting the scope of this disclosure.

[0005] Figure 1 It is a diagrammatic representation of a networked environment in which the content of this disclosure can be deployed, based on some examples.

[0006] Figure 2 It is a graphical representation of a messaging system with both client-side and server-side functionalities, based on some examples.

[0007] Figure 3 It is a graphical representation based on examples such as data structures maintained in a database.

[0008] Figure 4 It is a graphical representation based on some example messages.

[0009] Figure 5 These are stereoscopic diagrams based on some examples of eyeglass devices.

[0010] Figure 6 This is a flowchart illustrating an example method of a pupil center shift correction system based on some examples.

[0011] Figures 7 to 8 This is an illustrative component based on some examples of pupil center shift correction systems.

[0012] Figure 9 This is an illustrative screen based on some examples of pupil center shift correction systems.

[0013] Figure 10 It is a graphical representation of a machine in the form of a computer system, based on some examples, in which a set of instructions can be executed to cause the machine to perform any or more of the methods discussed herein.

[0014] Figure 11 It is a block diagram showing a software architecture in which examples can be implemented. Detailed Implementation

[0015] The following description discusses illustrative examples of the contents of this disclosure. In the description below, numerous specific details are set forth for illustrative purposes to provide an understanding of various examples of the disclosed subject matter. However, it will be apparent to those skilled in the art that examples of the disclosed subject matter can be practiced without these specific details. Generally, it is not necessary to show in detail well-known examples of instructions, protocols, structures, and techniques.

[0016] Typical smart glasses platforms enable users to interact with various types of virtual content. Such platforms are configured to display virtual content through the lenses of the smart glasses onto the real-world environment seen through those lenses. To interact with the virtual content, smart glasses can include embedded sensors (e.g., cameras or video sensors) that track eye movements and pupil diameter. Based on the user's gaze position, the smart glasses can control the virtual content superimposed on the display or other content presented to the user. This allows users to interact with the content simply by looking in a certain direction (or focusing their attention on the virtual content). Some smart glasses can use embedded sensors to detect gestural input and then update the display of the virtual content. If test conditions do not match calibration conditions, interactions using eye gaze to perform various modifications to the virtual content via embedded sensors may be inaccurate, potentially leading to various errors and miscalculations. Consequently, users cannot accurately perform the various expected interactions with the virtual content, detracting from the overall user experience.

[0017] Specifically, to calculate fixation (or gaze direction), video-based eye trackers calculate vectors between the pupil center and corneal reflexes to map how these vectors correspond to fixation position / direction. This relies on the assumption that the pupil center position is fixed. However, when the pupil constricts or dilates due to changes in brightness, accommodation, cognitive load, or arousal, the pupil center position can shift towards the nose and / or vertically, and the magnitude of this shift can vary from person to person. This can lead to errors in the estimation of fixation position / direction and convergence depth. To remedy this, some eye-tracking systems recommend that users calibrate using a screen with similar intensity / color / stimuli depth that will be present during the testing phase to reduce the likelihood of pupil center shift. This calibration can be used in settings where rigorous experimental control is possible. However, for any practical use in indoor and outdoor settings where brightness, stimulus depth, and arousal / emotional state levels may change between training and testing, pupil center shift will result in errors in fixation direction / position and convergence depth.

[0018] Some systems address these issues by estimating a corrected pupil center position using variations in brightness. These systems can provide correction using second-order polynomials, but such correction addresses the difference between the pupil center and corneal reflex. End users may not have access to these factors. Other systems perform calibration to induce changes in pupil size at different brightness values ​​and provide a device using a lookup table that maps pupil size to drift in the gaze position to correct for gaze position. However, such methods only provide correction for gaze position using a 2D eye tracker, not for changes in gaze direction provided by a 3D motion eye tracker. While existing corrections for 2D gaze position can correct for spatial accuracy, these corrections do not account for the convergence depth estimation error that also occurs. Furthermore, independent 2D eye-tracking methods exist that consider convergence depth estimation but cannot simultaneously account for both types of errors, as addressed by the disclosed example. That is, the disclosed example considers both convergence depth estimation and spatial accuracy.

[0019] The disclosed examples improve the efficiency of using electronic devices by providing AR devices (e.g., glasses devices) that enable users to interact with virtual content or AR objects displayed by the AR device by accurately tracking the user's gaze direction. To this end, the disclosed techniques train a machine learning model based on various pupil sizes and gaze eccentricities associated with the user to generate corrections to the gaze vector. This allows the disclosed techniques to accurately control virtual content based on gaze across the full range of pupil constriction and dilation. That is, the disclosed techniques correct the gaze vector (e.g., gaze direction) rather than the gaze location, which directly corrects both spatial accuracy and convergence depth. Using a regression-based approach, the disclosed techniques take into account additional nonlinearities and interactivity in the predictor (e.g., neural networks, tree-based methods) and provide a model that improves the capture of the relationship between pupil size and gaze prediction error.

[0020] Specifically, the disclosed technique accesses information identifying a point in the user's pupil (e.g., the pupil center) and an uncorrected gaze vector calculated based on that point to generate a correction to the uncorrected gaze vector. To this end, the disclosed technique processes an image of a pupil with a given diameter and gaze eccentricity using a machine learning model to predict estimation errors in the gaze vector. The machine learning model is trained to establish relationships between multiple ground-value gaze vectors and the uncorrected gaze vector for multiple pupil parameters such as pupil diameter, gaze angle, and / or gaze eccentricity. The disclosed technique generates a corrected gaze vector by applying the estimation errors in the gaze vector predicted by the machine learning model to the uncorrected gaze vector.

[0021] In this way, the disclosed examples increase the efficiency of electronic devices by reducing the amount of input and information required to complete the task and by reducing the amount of complex image processing algorithms running on AR devices. The disclosed examples also improve the efficiency, appeal, and usability of electronic AR devices such as glasses. While the disclosed examples are provided in the context of electronic glasses devices, similar examples can be applied to any other type of AR wearable device, such as AR hats, AR watches, AR belts, AR rings, AR bracelets, AR earrings, and / or AR headphones, or other devices that enable users to control or interact with content based on eye tracking or gaze direction.

[0022] Networked computing environment

[0023] Figure 1This is a block diagram illustrating an example interactive system 100 for exchanging data (e.g., messages and associated content) over a network. The interactive system 100 includes multiple instances of client devices 102, each hosting several applications including an interactive client 104 and other external applications 109 (e.g., third-party applications). Each interactive client 104 is communicatively coupled via a network 112 (e.g., the Internet) to other instances of the interactive client 104 (e.g., hosted on corresponding other client devices 102), a messaging server system 108, and an external app server 110. The interactive client 104 can also communicate with locally hosted third-party applications 109 using an application programming interface (API). The interactive client 104 may include messaging applications.

[0024] In some examples, the interaction system 100 includes a glasses device 119, which, among other applications, hosts a pupil center shift correction system 107. While only one example of the glasses device 119 is shown, any number of glasses devices 119 may be included in the interaction system 100. The glasses device 119 may represent any type of AR device worn by a user, such as AR glasses, AR hats, AR watches, AR belts, AR rings, AR bracelets, AR earrings, and / or AR headphones. The glasses device 119 is communicatively coupled to a client device 102 via a network 112, which may include a direct connection via a dedicated short-range communication path, such as Bluetooth™ or WiFi. In some cases, the client device 102 may also be referred to as a user device or simply as a device.

[0025] The pupil center shift correction system 107 enables a user to interact with virtual content or AR objects displayed by the glasses device 119 based on the user's eye gaze direction and / or movement. The glasses device 119 can access or collect motion data, such as inertial measurement unit data and image data obtained from multiple cameras within the glasses device 119, and can collect images or videos of the user's eyes (including the pupils). The glasses device 119 can determine its 3D movement based on the motion data and can compute an uncorrected gaze vector based on pupil detection in the images or videos. In some cases, the glasses device 119 can apply one or more machine learning models to estimated gaze eccentricity and pupil diameter from pupil (and / or eye) images to predict or estimate errors in the gaze vector. In such cases, the glasses device 119 can apply the predicted or estimated errors to the uncorrected gaze vector to generate a corrected gaze vector. The glasses device 119 can then perform corresponding modifications to one or more displayed AR objects, such as controlling the avatar based on the corrected gaze vector.

[0026] Interactive client 104 can communicate and exchange data with other interactive clients 104, glasses device 119, and message transceiver server system 108 via network 112. The data exchanged between interactive clients 104 and between interactive client 104 and message transceiver server system 108 includes functions (e.g., commands to activate functions) and payload data (e.g., text, audio, video, or other multimedia data).

[0027] The message transceiver server system 108 provides server-side functionality to a specific interactive client 104 via network 112. While some functions of the interactive system 100 are described herein as being performed by the interactive client 104 or by the message transceiver server system 108, the location of certain functions within the interactive client 104 or the message transceiver server system 108 can be a design choice. For example, it may be technically preferred that certain technologies and functions are initially deployed within the message transceiver server system 108, but that technology and functions are later migrated to the interactive client 104 of the client device 102, which has sufficient processing power.

[0028] The message transceiver server system 108 supports various services and operations provided to the interactive client 104. Such operations include sending data to and receiving data from the interactive client 104, and processing data generated by the interactive client 104. This data may include, for example, message content, client device information, geolocation information, media enhancements and overlays, message content persistence conditions, social network information, and on-site event information. Data exchange within the interactive system 100 is triggered and controlled via functions available through the user interface (UI) of the interactive client 104.

[0029] Specifically, turning to message transceiver server system 108, application programming interface (API) server 116 is coupled to application server 114 and provides a programming interface to application server 114. Application server 114 is communicatively coupled to database server 120, which facilitates access to database 126, which stores data associated with the messages processed by application server 114. Similarly, web server 128 is coupled to application server 114 and provides a web-based interface to application server 114. To this end, web server 128 handles incoming network requests via Hypertext Transfer Protocol (HTTP) and several other related protocols.

[0030] API server 116 receives and sends message data (e.g., commands and message payloads) between client device 102 and application server 114. Specifically, API server 116 provides a set of interfaces (e.g., routines and protocols) that can be invoked or queried by interactive client 104 to activate the functionality of application server 114. API server 116 exposes various functions supported by application server 114, including: account registration; login functionality; sending messages from one interactive client 104 to another interactive client 104 via application server 114; sending media files (e.g., images or videos) from interactive client 104 to message transceiver server 118 and for possible access by another interactive client 104; setting up collections of media data (e.g., stories); retrieving the friend list of the user of client device 102; retrieving such collections; retrieving messages and content; adding and deleting entities (e.g., friends) in an entity graph (e.g., a social graph); locating friends in the social graph; and opening (e.g., application events associated with interactive client 104).

[0031] Application server 114 hosts several server applications and subsystems, including, for example, messaging server 118, image processing server 122, and social networking server 124. Messaging server 118 implements several messaging techniques and functions, which specifically relate to the aggregation and other processing of content (e.g., text and multimedia content) included in messages received from multiple instances of interactive client 104. As will be described in further detail, text and media content from multiple sources can be aggregated into content collections (e.g., referred to as stories or galleries). These collections are then made available to interactive client 104. Such processing can also be performed on the server side by messaging server 118, taking into account the hardware requirements for additional processor- and memory-intensive processing of data.

[0032] Application server 114 also includes image processing server 122, which is dedicated to performing various image processing operations, typically relative to the payload of a message sent from or received at message transceiver server 118.

[0033] Image processing server 122 is used to implement the scanning function of augmented reality system 208. The scanning function includes activating and providing one or more AR experiences on client device 102 when an image is captured by client device 102. Specifically, interactive client 104 on client device 102 can be used to activate a camera device. The camera device displays one or more real-time images or videos and one or more icons or identifiers for one or more AR experiences to the user. As discussed above and below, the user can select a given identifier to initiate the corresponding augmented reality experience, for example, based on the user's pupil gaze direction. For example, the gaze direction can be determined as focusing or pointing at a specific AR experience within a threshold time period. In response, a specific AR experience can be initiated. Initiating an AR experience involves obtaining one or more augmented reality items associated with the AR experience and overlaying the augmented reality items onto the image or video being presented.

[0034] Social network server 124 supports various social networking functions and services and makes these functions and services available to messaging server 118. To this end, social network server 124 maintains and accesses entity graph 308 within database 126 (such as...). Figure 3 (As shown). Examples of the functions and services supported by the social network server 124 include identifying other users in the interaction system 100 who have a relationship with or are “following” that particular user, as well as identifying the interests and other entities of that particular user.

[0035] Returning to interactive client 104, the features and functionalities of the external resource (e.g., a third-party application or a mini-program) are made available to the user via the interface of interactive client 104. Interactive client 104 receives user selections regarding options for launching or accessing features of the external resource (e.g., a third-party resource, such as external app 109). The external resource may be a third-party application (external app 109) installed on client device 102 (e.g., a “native app”), or a smaller version (e.g., a “mini-program”) of a third-party application hosted on client device 102 or located remotely on client device 102 (e.g., on external application server 110). The smaller version of the third-party application includes a subset of the features and functionalities of the third-party application (e.g., a full, native version of a third-party standalone application) and is implemented using markup language documentation. In one example, the smaller version of the third-party application (e.g., a “mini-program”) is a web-based markup language version of the third-party application and is embedded in interactive client 104. In addition to using markup language documentation (e.g., ...), In addition to files, mini-programs can also incorporate scripting languages ​​(e.g., ... (files or .json files) and stylesheets (e.g., document).

[0036] In response to a user selection of an option to launch or access an external resource (external app 109), the interactive client 104 determines whether the selected external resource is a web-based external resource or a locally installed external application. In some cases, the external application 109, locally installed on the client device 102, can be launched independently of and separately from the interactive client 104, for example, by selecting an icon corresponding to the external application 109 on the client device 102's home screen (e.g., based on the user's pupil gaze direction). A smaller version of such an external application can be launched or accessed via the interactive client 104, and in some examples, no part of the smaller external application can be accessed outside the interactive client 104, or only a limited portion of the smaller external application can be accessed outside the interactive client 104. The smaller external application can be launched by receiving and processing a markup language document associated with the smaller external application from the external app server 110 via the interactive client 104.

[0037] In response to determining that the external resource is a locally installed external application 109, the interactive client 104 instructs the client device 102 to launch the external application 109 by executing locally stored code corresponding to the external application 109. In response to determining that the external resource is a web-based resource, the interactive client 104 communicates with the external app server 110 to obtain a markup language document corresponding to the selected resource. The interactive client 104 then processes the obtained markup language document to render the web-based external resource within the user interface of the interactive client 104.

[0038] Interactive client 104 can notify users of client device 102 or other users (e.g., "friends") associated with such users of one or more external resources. For example, interactive client 104 can provide participants in a conversation (e.g., a chat session) within interactive client 104 with notifications related to the current or recent use of an external resource by one or more members of a group of users. One or more users can be invited to join an active external resource or to activate an external resource that has been recently used but is currently inactive (in a group of friends). External resources can provide participants in the conversation (each using a corresponding client interactive client 104 with specific pupil characteristics) with the ability to share items, conditions, states, or locations within the external resource with one or more members of the user group in the chat session. Shared items can be interactive chat cards, which chat members can use to interact, for example, activate the corresponding external resource, view specific information within the external resource, or take a chat member to a specific location or state within the external resource (e.g., based on the user's pupil gaze direction). Within a given external resource, response messages can be sent to the user on interactive client 104. External resources can selectively include different media items in the response based on the current context of the external resource.

[0039] Interactive client 104 can present a list of available external resources (e.g., third-party or external applications 109 or mini-programs) to the user to launch or access a given external resource (e.g., based on the user's pupil gaze direction). This list can be presented in a context-sensitive menu. For example, icons representing different external applications in external application 109 (or mini-program) can vary based on how the user launches the menu (e.g., based on the user's pupil gaze direction). Any functions performed by pupil center shift correction system 107 can similarly be performed by interactive client 104.

[0040] System Architecture

[0041] Figure 2 This is a block diagram illustrating further details of an interactive system 100 based on some examples. Specifically, the interactive system 100 is shown as including an interactive client 104 and an application server 114. The interactive system 100 includes several subsystems supported on the client side by the interactive client 104 and on the server side by the application server 114. These subsystems include, for example, a short-lived timer system 202, a collection management system 204, an enhancement system 208, a map system 210, a game system 212, and an external resource system 220.

[0042] The short-lived timer system 202 is responsible for enabling temporary or time-limited access to content by the interactive client 104 and the message transceiver server 118. The short-lived timer system 202 includes several timers that selectively enable access (e.g., for rendering and displaying) of messages and associated content via the interactive client 104 based on the duration and display parameters associated with the message or message set (e.g., a story). Further details regarding the operation of the short-lived timer system 202 are provided below.

[0043] The collection management system 204 is responsible for managing groups and collections of media (e.g., collections of text, images, video, and audio data). Collections of content (e.g., messages including images, videos, text, and audio) can be organized into "event galleries" or "event stories." Such collections can be made available for a specified time period, such as the duration of an event related to the content. For example, content related to a concert can be made available as a "story" during the duration of the concert. The collection management system 204 can also be responsible for publishing icons that notify the user interface of the interactive client 104 of the existence of a specific collection.

[0044] Furthermore, the collection management system 204 includes a curation interface 206, which enables the collection manager to manage and curate specific content collections. For example, the curation interface 206 allows an event organizer to curate content collections related to a specific event (e.g., removing inappropriate content or redundant messages). Additionally, the collection management system 204 employs machine vision (or image recognition technology) and content rules to automatically curate content collections. In some examples, users may be paid compensation for including user-generated content in the collection. In such cases, the collection management system 204 operates to automatically pay such users for using their content.

[0045] Enhancement system 208 provides various functionalities that enable users to enhance (e.g., annotate or otherwise modify or edit) media content associated with a message. For example, enhancement system 208 provides functionalities related to generating and publishing media overlays for messages processed by interaction system 100. Enhancement system 208 operatively supplies media overlays or enhancements (e.g., image filters) to interactive client 104 based on the geolocation of client device 102. In another example, enhancement system 208 operatively supplies media overlays to interactive client 104 based on other information such as the social network information of the user of client device 102. Media overlays may include audio and visual content and visual effects. Examples of audio and visual content include images, text, logos, animations, and sound effects. Examples of visual effects include color overlays. Audio and visual content or visual effects may be applied to media content items (e.g., photographs) at client device 102. For example, media overlays may include text, graphic elements, or images that can be overlaid on photographs taken by client device 102. In another example, media overlays include location identifier overlays (e.g., Venice Beach), live event names, or business name overlays (e.g., beach cafes). In another example, enhancement system 208 uses the geolocation of client device 102 to identify a media overlay that includes the name of a merchant at the geolocation of client device 102. The media overlay may include additional tags associated with the merchant. The media overlay may be stored in database 126 and accessed through database server 120.

[0046] In some examples, enhancement system 208 provides a user-based publishing platform that allows users to select geolocations on a map and upload content associated with those geolocations. Users can also specify the environment in which specific media overlays should be provided to other users. Enhancement system 208 generates a media overlay that includes the uploaded content and associates the uploaded content with the selected geolocation.

[0047] In other examples, augmentation system 208 provides a merchant-based publishing platform that enables merchants to select specific media overlays associated with geolocation via a bidding process. For example, augmentation system 208 associates the media overlay of the highest bidder with a corresponding geolocation for a predefined amount of time. Augmentation system 208 communicates with image processing server 122 to obtain augmented reality experiences and presents identifiers of such experiences in one or more user interfaces (e.g., as icons on live images or videos, or as thumbnails or icons in interfaces specifically for presenting augmented reality experiences). Once an augmented reality experience is selected, one or more images, videos, or augmented reality graphic elements are retrieved and presented as overlays on images or videos captured by client device 102. In some cases, the camera is switched to a front-facing view (e.g., the front-facing camera of client device 102 is activated in response to the activation of a specific augmented reality experience), and images from the front-facing camera of client device 102, rather than the rear-facing camera of client device 102, begin to be displayed on client device 102. One or more image, video, or augmented reality graphic elements are retrieved and rendered as an overlay on an image captured and displayed by the front-facing camera of the client device 102.

[0048] Map system 210 provides various geolocation functions and supports the presentation of map-based media content and messages by interactive client 104. For example, map system 210 enables the display (e.g., stored in profile data 316) of user icons or avatars on a map to indicate the current or past location of the user's "friends" in the context of the map, as well as media content (e.g., a collection of messages including photos and videos) generated by such friends. For example, on the map interface of interactive client 104, messages posted by a user from a specific geographic location to interactive system 100 can be displayed to the specific user's "friends" in the context of that specific location on the map. Users can also share their location and status information with other users of interactive system 100 via interactive client 104 (e.g., using an appropriate status avatar), where the location and status information is similarly displayed to selected users in the context of the map interface of interactive client 104.

[0049] Game system 212 provides various game functions within the context of interactive client 104. Interactive client 104 provides a game interface that offers a list of available games (e.g., web-based games or web-based applications) that can be launched by a user within the context of interactive client 104 and played with other users of interactive system 100. Interactive system 100 also enables specific users to invite other users to play specific games by sending invitations from interactive client 104. Interactive client 104 also supports sending and receiving both voice and text messages (e.g., chat) within the game context, provides leaderboards for the game, and also supports providing in-game rewards (e.g., game currency and items).

[0050] External resource system 220 provides interactive client 104 with an interface to communicate with external application server 110 to launch or access external resources. Each external resource (app) server 110 hosts, for example, markup language-based applications (e.g., HTML5) or smaller versions of external applications (e.g., games, utilities, payment, or ride-sharing applications outside of interactive client 104). Interactive client 104 can launch web-based resources (e.g., applications) by accessing HTML5 files from external resource (app) servers 110 associated with the web-based resource. In some examples, applications hosted by external resource servers 110 are programmed in JavaScript using a software development kit (SDK) provided by messaging server 118. The SDK includes application programming interfaces (APIs) with functionality that can be called or triggered by web-based applications. In some examples, messaging server 118 includes a JavaScript library that provides access to certain user data of interactive client 104 to a given third-party resource. HTML5 is used as an example technology for programming games, but applications and resources programmed based on other technologies can be used.

[0051] To integrate the functionality of the SDK into a web-based resource, the SDK is downloaded by the external resource (app) server 110 from the messaging server 118 or otherwise received by the external resource (app) server 110. Once downloaded or received, the SDK is included as part of the application code of the web-based external resource. The code of the web-based resource can then call or activate certain functions of the SDK to integrate the features of the interactive client 104 into the web-based resource.

[0052] The SDK stored on message transceiver server 118 effectively bridges the gap between external resources (e.g., third-party or external applications 109 or mini-programs) and interactive client 104. This provides users with a seamless experience communicating with other users on interactive client 104 while preserving the look and feel of interactive client 104. To bridge communication between external resources and interactive client 104, in some examples, the SDK facilitates communication between external resource (app) server 110 and interactive client 104. In some examples, a WebViewJavaScriptBridge running on client device 102 establishes two unidirectional communication channels between external resources and interactive client 104. Messages are sent asynchronously between external resources and interactive client 104 via these communication channels. Each SDK function call is sent as a message and callback. Each SDK function is implemented by constructing a unique callback identifier and sending a message with that callback identifier.

[0053] By using the SDK, not all information from the interactive client 104 is shared with the external resource (app) server 110. The SDK limits which information is shared based on the needs of the external resource. In some examples, each external resource (app) server 110 provides an HTML5 file corresponding to the web-based external resource to the messaging server 118. The messaging server 118 can add a visual representation (e.g., box art or other graphics) of the web-based external resource to the interactive client 104. Once the user selects a visual representation or instructs the interactive client 104 to access a feature of the web-based external resource through the GUI of the interactive client 104, the interactive client 104 obtains the HTML5 file and instantiates the resource required to access the feature of the web-based external resource.

[0054] Interactive client 104 presents a graphical user interface (GUI) of an external resource (e.g., a login page or title screen). During, before, or after presenting the login page or title screen, interactive client 104 determines whether the initiated external resource has previously been authorized to access the user data of interactive client 104. In response to determining that the initiated external resource has previously been authorized to access the user data of interactive client 104, interactive client 104 presents another GUI of the external resource, including the functionality and characteristics of the external resource. In response to determining that the initiated external resource has not previously been authorized to access the user data of interactive client 104, after displaying the login page or title screen of the external resource for a threshold time period (e.g., 3 seconds), interactive client 104 slides up a menu (e.g., animates the menu to appear from the bottom of the screen to the middle of the screen or other parts) to authorize the external resource to access the user data. This menu identifies the type of user data that the external resource will be authorized to use. In response to receiving a user selection of the accept option, interactive client 104 adds the external resource to the list of authorized external resources and enables the external resource to access the user data from interactive client 104. In some examples, external resources are authorized by the interactive client 104 to access user data according to the OAuth 2 framework.

[0055] Interactive client 104 controls the type of user data shared with external resources based on the type of authorized external resource. For example, external resources including full-scale external applications (e.g., third-party or external application 109) are granted access to a first type of user data (e.g., a two-dimensional avatar of a user, with or without different avatar characteristics). As another example, external resources including smaller versions of external applications (e.g., a web-based version of a third-party application) are granted access to a second type of user data (e.g., payment information, a user's two-dimensional avatar, a user's three-dimensional avatar, and avatars with various avatar characteristics). Avatar characteristics include different ways of customizing the appearance and feel of an avatar, such as different poses, facial features, clothing, etc.

[0056] Data Architecture

[0057] Figure 3 This is a schematic diagram illustrating a data structure 300 that can be stored in a database 126 of a message transceiver server system 108, according to certain examples. Although the contents of the database 126 are shown as including several tables, it should be understood that the data can be stored in other types of data structures (e.g., as an object-oriented database).

[0058] Database 126 includes message data stored in message table 302. For any given message, this message data includes at least message sender data, message receiver (or recipient) data, and payload. See below for reference. Figure 4 Further details are described regarding information that can be included in the message and in the message data stored in message table 302.

[0059] Entity table 306 stores entity data and (for example, links to entity diagram 308 and profile data 316). Entities whose records are maintained in entity table 306 can include individuals, companies, organizations, objects, locations, events, etc. Regardless of the entity type, any entity whose data is stored in message transceiver server system 108 can be an identifiable entity. Each entity is provided with a unique identifier and an entity type identifier (not shown).

[0060] Entity graph 308 stores information about the relationships and associations between entities. Such relationships may be, for example, social, professional (e.g., working in the same company or organization), interest-based, or activity-based.

[0061] Profile data 316 stores various types of profile data about a specific entity. Based on privacy settings specified by the specific entity, profile data 316 can be selectively used and presented to other users of the interaction system 100. In the case of an individual, profile data 316 includes, for example, a username, phone number, address, and settings (e.g., notification and privacy settings) as well as an avatar representation (or a set of such avatar representations) selected by the user. A specific user can then selectively include one or more of these avatar representations within the content of messages transmitted via the interaction system 100, and on a map interface displayed to other users through the messaging client 104. The set of avatar representations may include a "status avatar," which presents a graphical representation of a status or activity that the user can choose to communicate at a specific time.

[0062] In the case that the entity is a group, in addition to the group name, members and various settings of the associated group (e.g., notifications), the group profile data 316 may similarly include one or more avatars associated with the group.

[0063] Database 126 also stores enhancement data, such as overlays or filters, in enhancement table 310. The enhancement data is associated with and applied to videos (video data is stored in video table 304) and images (image data is stored in image table 312).

[0064] In one example, a filter is an overlay displayed on an image or video during presentation to the receiving user. Filters can be of various types, including filters selected by the user from a set of filters presented to the sending user by the interactive client 104 when the sending user composes a message. Other types of filters include geolocation filters (also known as geographic filters), which can be presented to the sending user based on geolocation. For example, geolocation filters specific to nearby or particular locations can be presented by the interactive client 104 within the user interface based on geolocation information determined by the Global Positioning System (GPS) unit of the client device 102.

[0065] Another type of filter is a data filter, which can be selectively presented to the sending user by the interactive client 104 based on other inputs or information collected by the client device 102 during message creation processing. Examples of data filters include the current temperature at a specific location, the current speed of the sending user, the battery life of the client device 102, or the current time.

[0066] Other augmented data that can be stored in image table 312 includes augmented reality content items (e.g., corresponding to an applied augmented reality experience). Augmented reality content items or augmented reality items can be real-time special effects and sounds that can be added to images or videos.

[0067] As described above, augmented data includes augmented reality content items, overlays, image transformations, AR images, and similar items referring to modifications that can be applied to image data (e.g., video or images). This includes real-time modifications, which modify an image as it is captured by the device sensors (e.g., one or more cameras) of client device 102 and then displayed on the screen of client device 102 in the case of modification. This also includes modifications to stored content (e.g., video clips in a modifiable gallery). For example, in client device 102 with access to multiple augmented reality content items, a user can use a single video clip with multiple augmented reality content items to see how different augmented reality content items will modify the stored clip. For example, by selecting different augmented reality content items for the same content, multiple augmented reality content items with different pseudo-random motion models can be applied to that same content. Similarly, real-time video capture can be used with the illustrated modifications to show how a video image currently captured by the sensors of client device 102 will modify the captured data. Such data can be simply displayed on the screen without being stored in memory, or the content captured by the device's sensors can be recorded and stored in memory with or without modification (or both). In some systems, preview features can show how different augmented reality content items will look simultaneously in different windows on the display. For example, this can make it possible to view multiple windows with different pseudo-random animations on the display at the same time.

[0068] Therefore, using augmented reality content items and various systems, or other such transformation systems that use that data to modify the content, can involve: detection of objects (e.g., faces, hands, bodies, cats, dogs, surfaces, objects, etc.); tracking of such objects as they leave, enter, and move around within the field of view of a video frame; and modification or transformation of such objects while they are being tracked. In various examples, different methods can be used to implement such transformations. Some examples may involve generating a 3D mesh model of one or more objects and using transformations of the model and animated textures within the video to implement the transformation. In other examples, tracking points on the object can be used to place an image or texture (which can be two-dimensional or three-dimensional) at the tracked location. In further examples, neural network analysis of video frames can be used to place images, models, or textures within the content (e.g., images or frames of a video). Therefore, augmented reality content items refer both to the images, models, and textures used to create transformations within the content and to the additional modeling and analysis information required to implement such transformations using object detection, tracking, and placement.

[0069] Real-time video processing can be performed using any type of video data (e.g., video streams, video files, etc.) stored in the memory of any type of computerized system. For example, a user can load a video file and store it in the device's memory, or the device's sensors can be used to generate a video stream. Furthermore, computer-animated models can be used to process any object, such as human faces and parts of the human body, animals, or inanimate objects (e.g., chairs, cars, or other objects).

[0070] In some examples, when a specific modification is selected along with the content to be transformed, the element to be transformed is identified by a computing device, and then, if the element to be transformed exists in a frame of the video, it is detected and tracked. The elements of the object are modified according to the modification request, thereby transforming the frames of the video stream. For different kinds of transformations, the transformation of the video stream frames can be performed using different methods. For example, for frame transformations that primarily refer to changes in the elements of an object, feature points of each element of the object are calculated (e.g., using an Active Shape Model (ASM) or other known methods). Then, a feature point-based mesh is generated for each of at least one element of the object. This mesh is used in subsequent stages of tracking the elements of the object in the video stream. During tracking, the mesh for each element is aligned with the position of each element. Then, additional points are generated on the mesh. A first set of first points is generated for each element based on the modification request, and a second set of points is generated for each element based on this first set of first points and the second set of second points, as well as the mesh, to transform the frames of the video stream. In such a method, the background of the modified object can also be altered or distorted by tracking and modifying the background.

[0071] In some examples, transformations that alter some regions of an object using its elements can be performed by calculating feature points for each element of the object and generating a mesh based on those calculated feature points. Points are generated on the mesh, and then various regions are generated based on those points. The elements of the object are then tracked by aligning the region of each element with the position of at least one of the elements, and the properties of the regions can be modified based on requests for modification, thereby transforming frames of the video stream. Depending on the specific request for modification, the properties of the mentioned region can be transformed in different ways. Such modifications can involve: changing the color of the region; removing at least some portions of the region from the frames of the video stream; including one or more new objects in the region based on the modification request; and modifying or distorting the elements of the region or object. In various examples, any combination of such modifications or other similar modifications can be used. For certain models to be animated, some feature points can be selected as control points to be used throughout the state space that determines the options used for model animation.

[0072] In some examples of computer animation models that use face detection to transform image data, a specific face detection algorithm (e.g., Viola-Jones) is used to detect faces on the image. The Active Shape Model (ASM) algorithm is then applied to the facial regions of the image to detect facial feature reference points.

[0073] Other methods and algorithms suitable for face detection can be used. For example, in some examples, landmarks are used to locate features that represent distinguishable points present in most of the images considered. For example, for a face landmark, the location of the left eye pupil could be used. If the initial landmark is unrecognizable (e.g., if the person is wearing an eye patch), secondary landmarks can be used. Such a landmark recognition process can be used for any such object. In some examples, a set of landmarks forms a shape. The shape can be represented as a vector using the coordinates of the points in the shape. One shape is aligned with another shape using a similarity transformation (allowing translation, scaling, and rotation) that minimizes the average Euclidean distance between the points of the shapes. The average shape is the average of the aligned training shapes.

[0074] In some examples, the search begins by finding landmarks from an average shape aligned with the position and size of a face determined by a global face detector. This search then repeats the steps of adjusting the localization of shape points to suggest provisional shapes through template matching of the image texture around each point, and then conforming the provisional shapes to a global shape model until convergence occurs. In some systems, individual template matching is unreliable, and the shape model pools the results of weak template matching to form a stronger overall classifier. The entire search is repeated at each level of the image pyramid, from coarse to fine resolution.

[0075] The transformation system can capture image or video streams on a client device (e.g., client device 102) and perform complex image manipulations locally on client device 102 while maintaining an appropriate user experience, computation time, and power consumption. Complex image manipulations can include size and shape changes, emotion transfers (e.g., changing a face from frowning to smiling), state transfers (e.g., aging a subject, reducing apparent age, or changing gender), style transfers, application of graphic elements, and any other suitable image or video manipulations implemented by a convolutional neural network that has been configured to execute efficiently on client device 102.

[0076] In some examples, a computer-animated model for transforming image data can be used by a system in which a user can capture an image or video stream (e.g., a selfie) using a client device 102 that operates as part of an interactive client 104 operating on client device 102. A transformation system operating within the interactive client 104 determines the presence of a face within the image or video stream and provides a modification icon associated with the computer-animated model to transform the image data, or the computer-animated model can be presented as associated with the interface described herein. The modification icon includes changes that can be the basis for modifying the user's face within the image or video stream as part of a modification operation. Once a modification icon is selected, the transformation system initiates a process of transforming the user's image to reflect the selected modification icon (e.g., generating a smiley face on the user). Once the image or video stream is captured and the specified modification is selected, the modified image or video stream can be presented in a graphical user interface displayed on client device 102. The transformation system can implement a complex convolutional neural network on a portion of the image or video stream to generate and apply the selected modification. In other words, users can capture image or video streams, and once an edit icon is selected, the changes can be presented in real-time or near real-time. Furthermore, the changes can be persistent while a video stream is being captured and the selected edit icon is being toggled. Machine-trained neural networks can be used to achieve such modifications.

[0077] A graphical user interface (GUI) presenting modifications performed by a transformation system can provide users with additional interactive options. Such options can be based on an interface used to initiate content capture and selection for a specific computer animation model (e.g., initiated from a content creator GUI). In various examples, modifications can be persistent after an initial selection of the modification icon. Users can toggle modifications on or off by tapping or otherwise selecting a face modified by the transformation system and save it for later viewing or browsing to other areas of the imaging application. In cases where multiple faces are modified by the transformation system, users can globally toggle modifications on or off by tapping or selecting a single face modified and displayed within the GUI. In some examples, individual faces within a set of multiple faces can be modified separately, or such modifications can be toggled individually by tapping or selecting individual faces or a series of individual faces displayed within the GUI.

[0078] Story table 314 stores data about collections of messages and associated image, video, or audio data, which are compiled into collections (e.g., stories or galleries). The creation of a specific collection can be initiated by a specific user (e.g., each user who maintains a record in entity table 306). A user can create a "personal story" in the form of a collection of content that has been created and sent / broadcast by that user. For this purpose, the user interface of interactive client 104 may include user-selectable icons that allow the sending user to add specific content to his or her personal story.

[0079] The collection can also constitute a "live story" as a collection of content from multiple users, created manually, automatically, or using a combination of manual and automatic techniques. For example, a "live story" can constitute a curated stream of user-submitted content from various locations and events. Users whose client devices have location services enabled and who are at a co-located event at a specific time can be presented with the option to contribute content to a specific live story, for example, via the user interface of interactive client 104. Live stories can be identified to a user by interactive client 104 based on their location. The end result is a "live story" told from a community perspective.

[0080] Another type of content collection is called a "location story," which allows users whose client device 102 is located within a specific geographic location (e.g., on a college or university campus) to contribute to a specific collection. In some examples, contributing to a location story may require a second level of authentication to verify that the end user belongs to a specific organization or other entity (e.g., is a student on a university campus).

[0081] As mentioned above, video table 304 stores video data, which, in one example, is associated with messages whose records are maintained within message table 302. Similarly, image table 312 stores image data associated with messages whose message data is stored in entity table 306. Entity table 306 can associate various enhancements from enhancement table 310 with various images and videos stored in image table 312 and video table 304.

[0082] Data communication architecture

[0083] Figure 4This is a schematic diagram illustrating the structure of message 400 according to some examples. Message 400 is generated by interactive client 104 for transmission to another interactive client 104 or message transceiver server 118. The content of a particular message 400 is used to populate message table 302 stored in database 126, which is accessible by message transceiver server 118. Similarly, the content of message 400 is stored in memory as "in transit" or "in flight" data of client device 102 or application server 114. Message 400 is shown to include the following example components:

[0084] Message Identifier 402: A unique identifier that identifies message 400;

[0085] Message text payload 404: Text to be generated by the user via the user interface of client device 102 and included in message 400;

[0086] Message image payload 406: Image data captured by the camera component of the client device 102 or retrieved from the memory component of the client device 102 and included in the message 400. The image data for the sent or received message 400 may be stored in the image table 312;

[0087] Message video payload 408: Video data captured by the camera device component or retrieved from the memory component of the client device 102 and included in message 400. The video data for the sent or received message 400 can be stored in video table 304;

[0088] Message audio payload 410: Audio data captured by the microphone or retrieved from the memory component of the client device 102 and included in message 400;

[0089] Message enhancement data 412: Enhancement data (e.g., filters, stickers, or other annotations or enhancements) representing enhancements to be applied to the message image payload 406, message video payload 408, or message audio payload 410 of message 400. Enhancement data for the sent or received message 400 can be stored in enhancement table 310;

[0090] Message duration parameter 414: A parameter value, in seconds, indicating the amount of time that the content of the message (e.g., message image payload 406, message video payload 408, message audio payload 410) will be presented to the user via the interactive client 104 or that is accessible to the user.

[0091] Message geolocation parameter 416: Geolocation data (e.g., latitude and longitude coordinates) associated with the content payload of the message. Multiple message geolocation parameter 416 values ​​may be included in the payload, each of which is associated with a content item included in the content (e.g., a specific image in the message image payload 406, or a specific video in the message video payload 408);

[0092] Message Story Identifier 418: An identifier value that identifies one or more content sets (e.g., "Stories" identified in Story Table 314), wherein a specific content item in the message image payload 406 of message 400 is associated with one or more content sets. For example, the identifier value can be used to associate multiple images within message image payload 406 with multiple content sets, respectively;

[0093] Message Tag 420: Each message 400 can be labeled with multiple tags, each of which indicates the subject of the content included in the message payload. For example, in the case where a specific image depicts an animal (e.g., a lion) is included in the message image payload 406, the tag value can be included within the message tag 420 indicating the relevant animal. Tag values ​​can be generated manually based on user input, or they can be automatically generated using, for example, image recognition.

[0094] Message sender identifier 422: An identifier (e.g., a message sending system identifier, email address, or device identifier) ​​indicating the user of the client device 102 on which message 400 was generated and from which message 400 was sent; and

[0095] Message receiver identifier 424: An identifier (e.g., message sending and receiving system identifier, email address, or device identifier) ​​indicating the user of the client device 102 to which message 400 is addressed.

[0096] The content (e.g., values) of each component of message 400 can be pointers to locations in tables where content data values ​​are stored. For example, an image value in message image payload 406 can be a pointer to a location (or an address of a location) within image table 312. Similarly, a value in message video payload 408 can point to data stored in video table 304, a value in message enhancement data 412 can point to data stored in enhancement table 310, a value in message story identifier 418 can point to data stored in story table 314, and values ​​in message sender identifier 422 and message receiver identifier 424 can point to user records stored in entity table 306.

[0097] Eyeglasses equipment

[0098] Figure 5 A front perspective view of an eyeglass device 119 in the form of a pair of smart glasses including a pupil center shift correction system 107, according to some examples, is shown. The eyeglass device 119 includes a body 503 comprising a front or frame 506 and a pair of temples 509 connected to the frame 506 for supporting the frame 506 in a position on the user's face when the eyeglass device 119 is worn. The frame 506 can be made of any suitable material, such as plastic or metal (including shape memory alloys of any suitable shape). The frame 506 may include a touch input interface configured to receive touch input from the user (e.g., a one-finger touch, a two-finger touch, or a combination thereof, along with dragging a finger along the frame 506, such as a lateral end piece 521).

[0099] The eyewear device 119 includes a pair of optical elements in the form of a pair of lenses 512, which are held by corresponding optical element holders in the form of a pair of frame 515 forming part of a frame 506. The frame 515 are connected by a bridging member 518. In other examples, one or both of the optical elements may be a display, a display assembly, or a combination of lenses and a display.

[0100] Frame 506 includes a pair of end pieces 521 defining side ends of frame 506. In this example, various electronic components are housed in one or both end pieces 521. Temples 509 are coupled to the respective end pieces 521. In this example, temples 509 are coupled to frame 506 via corresponding hinges to be hingedly movable between a wearable mode and a folded mode, in which the temples 509 pivot toward frame 506 to lie substantially flat against frame 506. In other examples, temples 509 may be coupled to frame 506 by any suitable means, or may be rigidly or fixedly attached to frame 506 to become integral with it.

[0101] Each of the temples 509 includes a front portion coupled to the frame 506 and any suitable rear portion for coupling to the user's ear, such as... Figure 5 The curves shown in the examples. In some examples, frame 506 is formed from a single piece of material to have a monolithic or monolithic construction. In some examples, the entire body 503 (including both frame 506 and temple 509) can be a monolithic or monolithic construction.

[0102] The eyeglasses device 119 has onboard electronic components, including computing devices such as computer 524 or low-power processors, which can be of any suitable type in different examples for carrying by the body 503. In some examples, computer 524 is at least partially housed in one or both of the temples 509. In this example, various components of computer 524 are housed in the lateral endpieces 521 of frame 506. Computer 524 includes one or more processors, as well as memory (e.g., volatile storage devices, such as random access memory or registers), storage devices (e.g., non-volatile storage devices), wireless communication circuitry (e.g., BLE communication devices and / or WiFi Direct devices), and a power supply. Computer 524 includes low-power circuitry, high-speed circuitry, and in some examples, a display processor. Various examples may include these elements in different configurations or integrated in different ways.

[0103] The computer 524 additionally includes a battery 527 or other suitable portable power source. In one example, the battery 527 is disposed in one of the temples 509. Figure 5 In the eyewear device 119 shown, a battery 527 is shown disposed in one of the terminals 521 and electrically coupled to the rest of the computer 524 housed in the corresponding terminal 521.

[0104] The eyeglasses device 119 has camera functionality, and in this example, includes multiple camera devices 530 mounted in one of the end pieces 521 and facing forward, so as to be more or less aligned with the viewing direction of the wearer of the eyeglasses device 119. The multiple camera devices 530 are configured to capture digital images (also referred to herein as digital photographs or pictures) and digital video content. The operation of the multiple camera devices 530 is controlled by a camera device controller provided by a computer 524, meaning that image data of the images or videos captured by the multiple camera devices 530 is temporarily stored on a memory forming part of the computer 524. The multiple camera devices 530 may also include camera devices pointed at the user's face to capture images of the user's eyes. These images can be used to detect the position and size of the pupils of the eyes in order to estimate or generate a gaze vector representing the gaze direction of the eyes.

[0105] In some examples, virtual content is received from client device 102. In some examples, virtual content is received directly from application server 114. Onboard computer 524 receives input from the user, which drags or moves the avatar to a specific display position. The input may indicate whether the display position is anchored to a specific real-world object. In this case, when the lens 512 is moved to view different parts of the real-world environment, the avatar remains fixed in the display position of the specific real-world object, and if the lens 512 is rotated or moved a sufficient distance from the avatar's display position, the avatar can be removed from the field of view. In some examples, the display position is not anchored, in which case the avatar's display position is also updated to remain within the field of view when the lens 512 is moved to view different parts of the real-world environment. This allows the user to move around them and consistently and continuously see the avatars of the user and the user they are engaging in voice-based conversations with.

[0106] The glasses device 119 includes one or more inertial measurement units (IMUs), such as accelerometers and / or gyroscopes, and a touch interface. Based on input received by the glasses device 119 from the IMUs and / or touch interface, the glasses device 119 can control user interaction with virtual content. The IMUs can be used to determine the movement, rotation, velocity, direction, and orientation of the wearer's head to generate 3D motion information for the wearer's avatar or some other virtual objects included and displayed by the glasses device 119.

[0107] 3D motion information can represent head movement, facial feature movement, limb, joint, and / or other body movement. Motion information can be used by the eyewear device 119 to control one or more avatars presented on the lenses of the eyewear device 119. For example, if the motion information indicates movement along a specific direction and / or at a specific speed exceeding a threshold, such motion information can then be used to animate the displayed avatar as running or to animate the displayed avatar as running, instead of walking in the same specific direction. As another example, if the motion information indicates acceleration or vertical acceleration exceeding a threshold and vertical displacement or height exceeding a threshold amount, the motion information can then be used to animate the displayed avatar as jumping or bending over. As another example, if the motion information indicates a 360-degree turn along a specified curve, the motion information can then be used to animate the displayed avatar as rotating about an axis or to animate the displayed avatar as rotating about an axis, to, for example, simulate a somersault. 3D motion information can be combined with gaze vectors to precisely control and interact with the virtual content presented by the glasses device 119.

[0108] The glasses device 119 also includes one or more communication devices, such as a Bluetooth Low Energy (BLE) communication interface. This BLE communication interface enables the glasses device 119 to communicate wirelessly with the client device 102. Alternatively, or in addition to a BLE communication interface, other forms of wireless communication may be used, such as a WiFi direct connection. The BLE communication interface implements a standard number of BLE communication protocols.

[0109] A first communication protocol implemented by the BLE interface of the glasses device 119 enables the establishment of an unencrypted link between the glasses device 119 and the client device 102. In this first protocol, the link-layer communication (physical interface or medium) between the glasses device 119 and the client device 102 includes unencrypted data. In this first protocol, the application layer (the communication layer that operates on the physically exchanged data) encrypts and decrypts the data physically exchanged in unencrypted form through the link layer via the BLE communication interface. In this way, although the data exchanged through the physical layer can be freely read by an eavesdropping device, the eavesdropping device cannot decipher the exchanged data without performing decryption operations at the application layer.

[0110] A second communication protocol implemented by the BLE interface of the glasses device 119 enables the establishment of an encrypted link between the glasses device 119 and the client device 102. In this second protocol, the link-layer communication (physical interface) between the glasses device 119 and the client device 102 receives data from the application layer and adds a first type of encryption to the data before exchanging data via the physical medium. In this second protocol, the application layer (the communication layer that operates on the physically exchanged data) may or may not use the second type of encryption to encrypt and decrypt the data, which is physically exchanged in encrypted form via the link layer of the BLE communication interface using the first type of encryption. That is, the data can be encrypted first by the application layer and then further encrypted by the physical layer before being exchanged via the physical medium. After being exchanged via the physical medium, the data is then decrypted by the physical layer and then decrypted again by the application layer (e.g., using a different type of encryption). In this way, the data exchanged via the physical layer cannot be read by eavesdropping devices because the data is encrypted within the physical medium.

[0111] In some examples, client device 102 communicates with glasses device 119 using a first protocol and / or a second protocol to exchange images, videos, or virtual content between interactive client 104 and glasses device 119.

[0112] Pupil center shift correction system

[0113] Figure 6This is a flowchart illustrating an example method of executing method or process 600 by a pupil center shift correction system 107 according to some examples. Process 600 may be embodied as computer-readable instructions for execution by one or more processors, such that the operation of process 600 may be performed, in part or in whole, by functional components of pupil center shift correction system 107; therefore, process 600 is described below by way of example with reference to pupil center shift correction system 107. However, in other examples, at least some operations of process 600 may be deployed on various other hardware configurations. Therefore, process 600 is not intended to be limited to pupil center shift correction system 107, but may be implemented wholly or in part by any other component. Some or all operations of process 600 may be performed in parallel, out of order, or omitted entirely.

[0114] At operation 601, the pupil center shift correction system 107 accesses an uncorrected gaze vector that has been calculated based on the center point of the user's pupil. For example, the pupil center shift correction system 107 may communicate with another system component (e.g., a black box component) that detects or identifies the center point or gaze point of the pupil and generates an uncorrected gaze vector based on that center point or gaze point.

[0115] At operation 602, the pupil center shift correction system 107 processes the pupil image using a machine learning model to predict the estimation error in the gaze vector. As discussed above and below, a machine learning model can be trained to establish relationships between multiple ground truth gaze vectors and uncorrected gaze vectors for multiple pupil parameters, including diameter, gaze angle, or gaze eccentricity. The following section combines... Figure 7 The training operations performed by the machine learning model are discussed in detail.

[0116] At operation 603, as discussed above and below, the pupil center shift correction system 107 generates a corrected gaze vector by applying the estimation error in the gaze vector predicted by the machine learning model to an uncorrected gaze vector that has been calculated based on the center point of the user's pupil.

[0117] Specifically, the pupil center shift correction system 107 enables users to interact with virtual content or AR objects displayed by an AR device by accurately tracking the user's gaze direction. To this end, the pupil center shift correction system 107 trains a machine learning model based on various pupil sizes (diameters) and pupil eccentricities associated with the user to generate corrections to the gaze vector. This allows the disclosed technique to accurately control virtual content based on gaze across the full range of pupil constriction and dilation. That is, the pupil center shift correction system 107 corrects the gaze vector (e.g., gaze direction) rather than the gaze location, thereby directly and simultaneously correcting both spatial accuracy and convergence depth. Using a regression-based approach, the pupil center shift correction system 107 considers additional nonlinearities and interactivity in the predictor (e.g., neural networks, tree-based methods) and provides a model that improves the capture of the relationship between pupil size and gaze prediction error.

[0118] In some examples, the pupil center shift correction system 107 trains a machine learning model to predict errors in gaze vectors based on training data collected for a specific user or a group of users. To generate training data for training the machine learning model, the pupil center shift correction system 107 initiates a calibration phase. During the calibration phase, one or more stimuli are presented to the user on a display, and training data is collected, including uncorrected gaze vectors for multiple pupil diameters and pupil eccentricities, and multiple ground truth gaze vectors associated with the uncorrected gaze vectors for multiple pupil diameters and eccentricities (e.g., pupil parameters or characteristics).

[0119] In some examples, the pupil center shift correction system 107 adjusts the backlight intensity to a first value selected between 0.5 nits and 2000 nits. These intensity values ​​capture the expected range of pupil diameter under variable conditions. The pupil center shift correction system 107 presents one or more stimuli at one or more known locations based on the backlight intensity with the first value. This backlight intensity with the first value causes the user's pupil to dilate by a first amount and become a first shape or size. The pupil center shift correction system 107 can then communicate with another component to obtain or access a first trained uncorrected gaze vector. The first trained uncorrected gaze vector can be generated based on the center point of the pupil having the first diameter or shape. The pupil center shift correction system 107 can then calculate a first known gaze vector based on the correlation between the first pupil diameter, gaze eccentricity, or shape and one or more known locations of one or more stimuli. Then, the pupil center shift correction system 107 can generate a first set of training data that associates the first pupil shape or size with a first uncorrected gaze vector and a first known gaze vector and / or gaze eccentricity.

[0120] Next, the pupil center shift correction system 107 adjusts the backlight intensity to a second value selected between 0.5 nits and 2000 nits. The pupil center shift correction system 107 presents one or more stimuli at one or more known locations based on the backlight intensity with the second value. This backlight intensity with the second value causes the user's pupil to dilate (or constrict) by a second amount and change to a second shape or size. The second value of the backlight intensity can be greater than or less than the first value. In some cases, the second value is greater than the first value, causing the backlight intensity to change from dark to bright, because pupil constriction due to pupillary light reflection may occur significantly faster than from bright to dark.

[0121] The pupil center shift correction system 107 can then determine a second training pupil diameter or shape based on an image of the pupil. The pupil center shift correction system 107 can communicate with external components to obtain a second training uncorrected gaze vector for the pupil now having the second pupil diameter or shape. The pupil center shift correction system 107 can then compute a second known gaze vector based on the correlation between the second pupil diameter or shape and one or more known locations of one or more stimuli. The pupil center shift correction system 107 can then generate a second set of training data that correlates the second pupil shape or size and location (gaze eccentricity) with the second training uncorrected gaze vector and the second known gaze vector. This process is repeated at various backlight intensity values ​​ranging from 0.5 nits to 2000 nits and for various stimulus locations. Once a specified amount of training data has been collected in a similar manner, the pupil center shift correction system 107 performs a training phase on its machine learning model.

[0122] Figure 7Illustrative components of a training phase 700 of a pupil center shift correction system 107 according to some examples are shown. The training phase 700 trains a machine learning model of the pupil center shift correction system 107 based on certain pupil sizes, eccentricities, shapes, and / or diameters to estimate or predict errors in uncorrected gaze vectors. The machine learning model of the pupil center shift correction system 107 may include multiple components, each associated with a different eye of the user (e.g., left / right eye or pupil). That is, because errors can differ across eyes, the model can be used to separately calculate the errors of the X-axis, Y-axis, and Z-axis angular components of the left and right gaze vectors. Each component of the machine learning model can be trained individually and independently to predict or estimate the corresponding angle error in a specified angle component (e.g., a first set of angle corrections including a first component can estimate the angle error of the left eye along the X-axis angle component, a second component can estimate the angle error of the left eye along the Y-axis angle component, and a third component can estimate the angle error of the left eye along the Z-axis angle component; and a second set of angle corrections including a first component can estimate the angle error of the right eye along the X-axis angle component, a second component can estimate the angle error of the right eye along the Y-axis angle component, and a third component can estimate the angle error of the right eye along the Z-axis angle component).

[0123] The machine learning model of the pupil center shift correction system 107 can be trained using labeled training data that includes ground truth information. For example, uncorrected gaze vectors for various pupil diameters (or shapes) can be included in the training data along with ground truth gaze vectors for the corresponding pupil diameters (shapes). The machine learning model can be applied to a subset of the training data, such as an initial batch of training data including a first uncorrected gaze vector 740 for a first pupil diameter 710, and can generate an estimate or prediction including a first estimation error 742 in the first uncorrected gaze vector 740. The first uncorrected gaze vector 740 can be calculated or generated based on the gaze origin 720. The first estimation error 742 can represent an angular shift 760 of the first uncorrected gaze vector 740 from the gaze origin 720 of the pupil (which can be measured relative to the center of the pupil or any other arbitrary point). That is, the gaze origin 720 can be used to generate the first uncorrected gaze vector 740 along a visual axis that intersects with the pupil center or the gaze origin 720.

[0124] The machine learning model can calculate the ground truth error between a first uncorrected gaze vector 740 and a first ground truth gaze vector 730 associated with the first uncorrected gaze vector 740. The machine learning model can then calculate the deviation between the ground truth error and a first estimation error. This deviation can then be used to update one or more parameters of the machine learning model. In some cases, one or more parameters of the machine learning model can be further updated based on gaze eccentricity data 750 associated with the initial training data. After updating one or more parameters, the machine learning model is applied to another subset of the training data, and these operations are repeated until a stopping criterion is met. This allows the glasses device 119 to be trained to accurately calculate / determine gaze orientation across different pupil shapes / sizes / diameters.

[0125] Machine learning is the field of research that gives computers the ability to learn without being explicitly programmed. Machine learning explores the research and construction of algorithms (also referred to as tools in this paper) that can learn from existing data and make predictions on new data. Such machine learning tools operate by building models based on example training data to make data-driven predictions or decisions that are represented as outputs or evaluations. While examples of several machine learning tools are given, the principles presented in this paper can be applied to other machine learning tools.

[0126] In some examples, different machine learning tools can be used. For instance, logistic regression (LR), Naive Bayes, random forest (RF), neural network (NN), matrix factorization, and support vector machine (SVM) tools can be used to classify or score job positions.

[0127] Two common problem types in machine learning are classification and regression problems. Classification problems (also known as categorization problems) aim to classify an item into one of several category values ​​(e.g., is the object an apple or an orange?). Regression algorithms aim to quantify some items (e.g., by providing values ​​as real numbers). Machine learning algorithms use features to analyze data to generate evaluations. Each feature is a separate, measurable property of the observed phenomenon. The concept of a feature is related to the concept of explanatory variables used in statistical techniques such as linear regression. Selecting features that provide useful information, are identifiable, and are independent is crucial for the effective operation of MLPs in pattern recognition, classification, and regression. Features can be of different types, such as numeric features, strings, and graphs.

[0128] In one example, and for illustrative purposes only, features can be of different types and can include one or more of content, concepts, attributes, historical data, and / or user data. Machine learning algorithms use training data to find correlations between identified features that influence outcomes or evaluations. In some examples, training data includes labeled data, which is known data about one or more identified features and one or more outcomes, such as detecting communication patterns, detecting the meaning of messages, generating message summaries, detecting actions in messages, detecting urgency in messages, detecting the relationship between users and senders, calculating score attributes, calculating message scores, detecting errors in uncorrected gaze vectors, etc.

[0129] Using training data and labeled features, a machine learning tool is trained at the machine learning program training stage. The machine learning tool evaluates the values ​​of features as they relate to the training data. The result of training is a trained machine learning program. When performing evaluations using the trained machine learning program, new data is fed as input, and the trained machine learning program generates evaluations as output.

[0130] Machine learning programs support two types of phases: a training phase and a prediction phase. In the training phase, supervised learning, unsupervised learning, or reinforcement learning can be used. For example, a machine learning program (1) receives features (e.g., structured or labeled data as in supervised learning) and / or (2) identifies features in the training data (e.g., unstructured or unlabeled data for unsupervised learning). In the prediction phase, the machine learning program uses the features to analyze the query data to generate results or predictions, as an example of evaluation.

[0131] During the training phase, feature engineering is used to identify features and may include identifying informative, discriminative, and independent features for efficient operation of machine learning programs in pattern recognition, classification, and regression. In some examples, the training data includes labeled data, which are known data of pre-identified features and one or more outcomes. Each of the features can be a variable or attribute, such as a measurable property of a process, item, system, or phenomenon represented by a dataset (e.g., the training data).

[0132] During the training phase, the machine learning program uses training data to find correlations between features that influence predictions or evaluations. The machine learning program is trained using the training data and the identified features during the training phase. The machine learning program evaluates the values ​​of features when they are correlated with the training data. The result of training is a trained machine learning program (e.g., a trained or learned model).

[0133] Furthermore, the training phase can involve machine learning, where the training data is structured (e.g., labeled during preprocessing), and the trained machine learning program implements a relatively simple neural network capable of performing, for example, classification and clustering operations. In other examples, the training phase can involve deep learning, where the training data is unstructured, and the trained machine learning program implements a deep neural network capable of performing both feature extraction and classification / clustering operations.

[0134] Neural networks generated during the training phase and implemented within a trained machine learning program can include a hierarchical (e.g., layered) organization of neurons. For example, neurons (or nodes) can be arranged hierarchically into several layers, including an input layer, an output layer, and multiple hidden layers. Each layer within the neural network can have one or more neurons, and each of these neurons can operationally compute a small function (e.g., an activation function). For example, if the activation function produces a result exceeding a certain threshold, the output can be passed from that neuron (e.g., a sending neuron) to connected neurons (e.g., receiving neurons) in subsequent layers. The connections between neurons also have associated weights that define the influence of the input from the sending neuron to the receiving neuron.

[0135] In some examples, a neural network can also be one of several different types of neural networks, including: single-layer feedforward networks, artificial neural networks (ANNs), recurrent neural networks (RNNs), symmetric connected neural networks and unsupervised pre-trained networks, convolutional neural networks (CNNs), generative adversarial networks (GANs) and / or recurrent neural networks (RNNs), to name just a few.

[0136] During the prediction phase, a trained machine learning program is used to perform the evaluation. Query data is fed as input to the trained machine learning program, and the trained machine learning program generates an evaluation as output in response to the received query data.

[0137] In some examples, the first component of the machine learning model is trained to predict or estimate the error in the gaze along the first angular component. In such cases, the machine learning model follows Equation 1: Δx model: (Xground_truth - Xuncorrected) ~ The deviation between the first uncorrected gaze vector 740 of the training data and the ground truth error between the first ground truth gaze vector 730 is calculated. That is, the first component updates the parameters of the machine learning model based on the deviation between the first angular component error generated by the first component of the machine learning model and the Δx output by Equation 1. The second component of the machine learning model is trained to predict or estimate the error in the gaze along the second angular component. In this case, the machine learning model follows Equation 2: Δy model: (Yground_truth - Yuncorrected) ~ The deviation between the first uncorrected gaze vector 740 of the training data and the ground truth error between the first ground truth gaze vector 730 is calculated. That is, the second component updates the parameters of the machine learning model based on the deviation between the second angular component error generated by the second component of the machine learning model and Δy output by Equation 2.

[0138] The third component of the machine learning model is trained to predict or estimate the gaze error along the third angular component. In this case, the machine learning model follows Equation 3: Δz model: (Zground_truth - Zuncorrected)~ The deviation between the first uncorrected gaze vector 740 of the training data and the ground truth error between the first ground truth gaze vector 730 is calculated. That is, the third component updates the parameters of the machine learning model based on the deviation between the third angular component error generated by the third component of the machine learning model and Δz output by Equation 3.

[0139] In this case, a machine learning model is trained separately for each of the x, y, and z angular components of the gaze vector and for each eye. In some examples, the ground truth vector (x, y, or z components) is compared with the x, y, or z components of the gaze vector (e.g., ... Figure 7 The simple difference between (as shown) can be the dependent variable in the model (Δx, Δy, Δz). Pupil diameter, the sum of squares of pupil diameter, and the reported fixation eccentricity (e.g., fixation angle) can be used as predictors in machine learning models. Regression can capture the interactivity and nonlinearity in the predictor to provide robust estimates of Δx, Δy, or Δz.

[0140] In some examples, after calibration and training, the machine learning model can be used as part of a pupil center shift correction system 107 to correct the gaze vector in real time. At any given time t, estimated Δx, Δy, and Δz can be calculated from the machine learning model and used to update the uncorrected gaze vector. For example, as... Figure 8As shown in Figure 800, the pupil center shift correction system 107 can communicate with external components to obtain access to the uncorrected gaze vector 820, which has been calculated based on the gaze origin 810 (which may correspond to the center point on the pupil), for example, based on a typical method for calculating the gaze vector using corneal reflection.

[0141] The pupil center shift correction system 107 can then use an imaging device to calculate or determine the size, shape, and / or diameter of the user's pupil using images of the pupil. The pupil center shift correction system 107 can apply a machine learning model to the size, shape, and / or diameter of the user's pupil to determine or predict an error 840 in the angular component of the gaze vector. This process can be performed individually for each of the user's eyes. The pupil center shift correction system 107 can then add the error 840 to the uncorrected gaze vector 820 to generate a predicted corrected gaze vector 830. The pupil center shift correction system 107 can use the predicted corrected gaze vector 830 to identify the object of interest to which the user's gaze is directed. Based on the identified object of interest, one or more interactions or modifications to the object of interest (e.g., a virtual object) can be performed.

[0142] For example, the glasses device 119 can present Figure 9 The user interface 900 shown includes a depiction of an avatar 932 (virtual object) in a real-world environment 910. When a user wears glasses device 119, glasses device 119 can use multiple onboard sensors of glasses device 119 to collect or determine pupil characteristics / parameters (e.g., pupil size, shape, diameter, gaze eccentricity, etc.). In response to determining the pupil characteristics / parameters, glasses device 119 calculates or estimates an error in the gaze direction or gaze vector. Glasses device 119 updates the uncorrected gaze vector generated based on a point on the pupil (e.g., the center point) based on the estimated error. The updated gaze vector can then be used to determine whether the user is gazing directly at avatar 932. If so, glasses device 119 can perform one or more operations relative to avatar 932, such as moving or positioning avatar 932 at a specified location or position in 3D space.

[0143] For example, an uncorrected gaze vector might point to virtual object 934. Without applying correction to the uncorrected gaze vector, the glasses device 119 might perform an operation relative to virtual object 934. The user might actually intend to perform the operation relative to avatar 932, not virtual object 934. The glasses device 119 can use a machine learning model (discussed above) to apply correction to the uncorrected gaze vector to generate a corrected gaze vector. By applying the correction, the glasses device 119 can determine that the corrected gaze vector is pointing towards avatar 932, not virtual object 934. The glasses device 119 can then perform or accurately represent the user's intent relative to avatar 932.

[0144] Example

[0145] Example 1. A method comprising: accessing an uncorrected gaze vector computed based on the center point of a user's pupil; processing an image of the pupil using a machine learning model to predict an estimation error in the gaze vector, the machine learning model being trained to establish relationships between multiple ground truth gaze vectors and the uncorrected gaze vector for multiple pupil parameters, including diameter, gaze angle, or gaze eccentricity; and generating a corrected gaze vector, which has been computed based on the center point of the user's pupil, by applying the estimation error in the gaze vector predicted by the machine learning model to the uncorrected gaze vector.

[0146] Example 2. According to the method of Example 1, it also includes: performing one or more augmented reality operations based on the corrected gaze vector.

[0147] Example 3. The method according to Example 2 further includes: identifying an object depicted in the display of the augmented reality device that corresponds to a calibrated gaze vector; and performing one or more augmented reality operations related to the identified object.

[0148] Example 4. According to any of Examples 1 to 3, wherein the machine learning model is also trained to establish a relationship between multiple ground truth gaze vectors and uncorrected gaze vectors based on gaze eccentricity.

[0149] Example 5. The method according to any one of Examples 1 to 4 further includes: training a first component of a machine learning model to predict a first estimation error in a first angular component; training a second component of a machine learning model to predict a second estimation error in a second angular component; and training a third component of a machine learning model to predict a third estimation error in a third angular component.

[0150] Example 6. Following the method of Example 5, the first, second, and third components are trained individually and independently of each other. These components can be computed independently for each eye or for the combined gaze vectors.

[0151] Example 7. According to any one of Examples 5 and 6, wherein the first part, the second part, and the third part are trained for each of the user's multiple eyes.

[0152] Example 8. The method according to any one of Examples 5 to 7 further includes: accessing multiple training data, the multiple training data including uncorrected gaze vectors for multiple pupil diameters and multiple ground truth gaze vectors associated with the uncorrected gaze vectors for multiple pupil diameters; obtaining a first batch of training data, the first batch of training data including a first uncorrected gaze vector for a first pupil diameter; processing the first pupil diameter with a machine learning model to predict a first estimation error in the first uncorrected gaze vector; calculating a ground truth error between the first uncorrected gaze vector and the first ground truth gaze vector associated with the first uncorrected gaze vector; calculating a bias / increment between the ground truth error and the first estimation error; and updating one or more parameters of the machine learning model based on the calculated bias.

[0153] Example 9. According to the method of Example 8, wherein the first estimation error includes multiple estimation errors for different angular components, the method further includes: calculating multiple true errors, each true error being associated with a different angular component of a first uncorrected gaze vector and a first true gaze vector; and calculating multiple deviations between the multiple true errors and the multiple estimation errors for the different angular components.

[0154] Example 10. Following the method of Example 9, one or more parameters of a machine learning model are updated based on multiple biases.

[0155] Example 11. The method of any one of Examples 8 to 10, wherein the true error is calculated based on the first pupil diameter, the square of the first pupil diameter, and the reported fixation eccentricity associated with the first pupil diameter.

[0156] Example 12. The method according to any one of Examples 8 to 11 further includes at least a portion of generating a plurality of training data by performing a calibration operation, the calibration operation including: presenting a stimulus on a display having a known location; determining a training pupil diameter; calculating a training uncorrected gaze vector based on the detected pupil diameter; calculating a known gaze vector based on the correlation between the detected pupil diameter and the known location of the stimulus; and storing the training uncorrected gaze vector and the detected pupil diameter associated with the known gaze vector as a plurality of training data.

[0157] Example 13. The method according to Example 12 further includes: repeating the calibration operation for multiple background light intensities.

[0158] Example 14. The method of Example 13, wherein multiple background light intensities are in the range of 0.5 nits to 2000 nits.

[0159] Example 15. The method of any one of Examples 1 to 14, wherein the uncorrected gaze vector is calculated based on the corneal reflection and the center of the pupil.

[0160] Example 16. The method according to any one of Examples 1 to 15 further includes detecting changes in the detection point of the pupil based on different environmental conditions, the changes in the detection point causing errors in the calculated uncorrected gaze vector, the errors including at least one of errors in gaze position, orientation, or convergence depth estimation.

[0161] Example 17. The method according to any one of Examples 1 to 16 further includes adding the estimated error in the gaze vector to the uncorrected gaze vector to generate a corrected gaze vector.

[0162] Example 18. A system comprising: at least one storage device; and at least one processor coupled to the at least one storage device and configured to perform operations including: accessing an uncorrected gaze vector computed based on the center point of a user's pupil; processing an image of the pupil using a machine learning model to predict an estimation error in the gaze vector, the machine learning model being trained to establish relationships between multiple ground truth gaze vectors and the uncorrected gaze vector for multiple pupil parameters, including diameter, gaze angle, or gaze eccentricity; and generating a corrected gaze vector, the uncorrected gaze vector having been computed based on the center point of the user's pupil, by applying the estimation error in the gaze vector predicted by the machine learning model to the uncorrected gaze vector.

[0163] Example 19. A system according to Example 18, wherein the operation includes: performing one or more augmented reality operations based on a corrected gaze vector.

[0164] Example 20. A non-transitory machine-readable storage medium comprising instructions that, when executed by one or more processors of the machine, cause the machine to perform operations including: accessing an uncorrected gaze vector computed based on the center point of a user's pupil; processing an image of the pupil using a machine learning model to predict an estimation error in the gaze vector, the machine learning model being trained to establish relationships between multiple ground truth gaze vectors and the uncorrected gaze vector for multiple pupil parameters, including diameter, gaze angle, or gaze eccentricity; and generating a corrected gaze vector, which has been computed based on the center point of the user's pupil, by applying the estimation error in the gaze vector predicted by the machine learning model to the uncorrected gaze vector.

[0165] Machine architecture

[0166] Figure 10 This is a schematic representation of machine 1000, within which instructions 1008 (e.g., software, programs, applications, applets, or other executable code) can be executed to cause machine 1000 to perform any or more of the methods discussed herein. For example, instructions 1008 can cause machine 1000 to perform any or more of the methods described herein. Instructions 1008 transform the general, non-programmed machine 1000 into a specific machine 1000 programmed to perform the described and illustrated functions in the described manner. Machine 1000 can operate as a standalone device or can be coupled (e.g., networked) to other machines. In a networked deployment, machine 1000 can operate as a server machine or a client machine in a server-client network environment, or as a peer machine in a peer-to-peer (or distributed) network environment. Machine 1000 may include, but is not limited to: server computers, client computers, personal computers (PCs), tablet computers, laptop computers, netbooks, set-top boxes (STBs), personal digital assistants (PDAs), entertainment media systems, cellular phones, smartphones, mobile devices, wearable devices (e.g., smartwatches), smart home devices (e.g., smart appliances), other smart devices, web devices, network routers, network switches, network bridges, or any machine capable of sequentially or otherwise executing instructions 1008 specifying actions to be taken by machine 1000. Furthermore, although only a single machine 1000 is shown, the term "machine" should also be considered as a collection of machines that individually or jointly execute instructions 1008 to perform any or more of the methods discussed herein. For example, machine 1000 may include client device 102 or any of several server devices forming part of message transceiver server system 108. In some examples, machine 1000 may also include both client and server systems, wherein certain operations of a particular method or algorithm are performed on the server side, and wherein certain operations of a particular method or algorithm are performed on the client side.

[0167] Machine 1000 may include a processor 1002, a memory 1004, and an input / output (I / O) unit 1038 that can be configured to communicate with each other via a bus 1040. In the example, processor 1002 (e.g., a central processing unit (CPU), a reduced instruction set computing (RISC) processor, a complex instruction set computing (CISC) processor, a graphics processing unit (GPU), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a radio frequency integrated circuit (RFIC), another processor, or any suitable combination thereof) may include, for example, processor 1006 and processor 1010 that execute instruction 1008. The term "processor" is intended to include multi-core processors, which may include two or more independent processors (sometimes referred to as "cores") capable of executing instructions simultaneously. Although Figure 10 Multiple processors 1002 are shown, but machine 1000 may include a single processor with a single core, a single processor with multiple cores (e.g., a multi-core processor), multiple processors with a single core, multiple processors with multiple cores, or any combination thereof.

[0168] Memory 1004 includes main memory 1012, static memory 1014, and storage cells 1016, all of which are accessible by processor 1002 via bus 1040. Main memory 1004, static memory 1014, and storage cells 1016 store instructions 1008 that implement any one or more of the methods or functions described herein. Instructions 1008 may also reside wholly or partially in main memory 1012, in static memory 1014, in machine-readable medium 1018 within storage cells 1016, in at least one of the processors 1002 (e.g., in the processor's cache memory), or in any suitable combination thereof during execution by machine 1000.

[0169] I / O component 1038 may include various components for receiving input, providing output, generating output, transmitting information, exchanging information, capturing measurement results, etc. The specific I / O component 1038 included in a particular machine will depend on the type of machine. For example, a portable machine such as a mobile phone may include a touch input device or other such input mechanism, while a headless server machine may not include such a touch input device. It should be understood that I / O component 1038 may include... Figure 10Many other components are not shown. In various examples, I / O component 1038 may include user output component 1024 and user input component 1026. User output component 1024 may include visual components (e.g., displays such as plasma display panels (PDPs), light-emitting diode (LED) displays, liquid crystal displays (LCDs), projectors, or cathode ray tube (CRT) displays), acoustic components (e.g., speakers), haptic components (e.g., vibration motors, resistance mechanisms), other signal generators, etc. User input component 1026 may include alphanumeric input components (e.g., keyboards, touchscreens configured to receive alphanumeric input, optical keyboards, or other alphanumeric input components), point-based input components (e.g., mice, touchpads, trackballs, joysticks, motion sensors, or other pointing instruments), haptic input components (e.g., physical buttons, touchscreens or other haptic input components that provide positioning and force for touch or touch gestures), audio input components (e.g., microphones), etc.

[0170] In another example, I / O component 1038 may include: biometric component 1028, motion component 1030, environmental component 1032, or position component 1034, and a variety of other components. For example, biometric component 1028 includes components for detecting expressions (e.g., hand expressions, facial expressions, voice expressions, body posture, or eye tracking), measuring biosignals (e.g., blood pressure, heart rate, body temperature, sweating, or brain waves), identifying people (e.g., voice identification, retinal identification, facial identification, fingerprint identification, or EEG-based identification), etc. Motion component 1030 includes accelerometer components (e.g., accelerometer), gravity sensor components, and rotation sensor components (e.g., gyroscope).

[0171] The environmental component 1032 includes, for example: one or more camera devices (with still image / photograph and video capabilities), lighting sensor components (e.g., photometers), temperature sensor components (e.g., one or more thermometers for detecting ambient temperature), humidity sensor components, pressure sensor components (e.g., barometers), acoustic sensor components (e.g., one or more microphones for detecting background noise), proximity sensor components (e.g., infrared sensors for detecting nearby objects), gas sensors (e.g., gas detection sensors for detecting the concentration of hazardous gases or measuring pollutants in the atmosphere for safety purposes), or other components that can provide indications, measurements, or signals corresponding to the surrounding physical environment.

[0172] Regarding the camera device, client device 102 may have a camera device system including, for example, a front-facing camera on the front surface of client device 102 and a rear-facing camera on the rear surface of client device 102. The front-facing camera may be used, for example, to capture still images and videos (e.g., “selfies”) of the user of client device 102, which can then be enhanced with the aforementioned enhancement data (e.g., filters). The rear-facing camera may be used, for example, to capture still images and videos in a more conventional camera device mode, which are similarly enhanced with the enhancement data. In addition to the front-facing and rear-facing cameras, client device 102 may also include a 360° camera for capturing 360° photos and videos.

[0173] Furthermore, the camera system of the client device 102 may include dual rear cameras (e.g., a main camera and a depth-sensing camera), or even include triple, quadruple, or quintuple rear camera configurations on the front and rear sides of the client device 102. For example, these multiple camera systems may include wide-angle cameras, ultra-wide-angle cameras, telephoto cameras, macro cameras, and depth sensors.

[0174] The position component 1034 includes a positioning sensor component (e.g., a GPS receiver component), an altitude sensor component (e.g., an altimeter or barometer that detects air pressure and can determine altitude based on air pressure), an orientation sensor component (e.g., a magnetometer), and the like.

[0175] Various technologies can be used to achieve communication. I / O component 1038 also includes communication component 1036, which is operable to couple machine 1000 to network 1020 or device 1022 via a corresponding coupling or connection. For example, communication component 1036 may include a network interface component or another suitable device that interfaces with network 1020. In further examples, communication component 1036 may include wired communication components, wireless communication components, cellular communication components, near field communication (NFC) components, Bluetooth® components (e.g., Bluetooth® Low Energy), Wi-Fi® components, and other communication components that provide communication via other modalities. Device 1022 may be another machine or any peripheral device from various peripheral devices (e.g., a peripheral device coupled via USB).

[0176] Furthermore, the communication component 1036 can detect identifiers, or include components operable to detect identifiers. For example, the communication component 1036 may include a radio frequency identification (RFID) tag reader component, an NFC smart tag detection component, an optical reader component (e.g., an optical sensor for detecting one-dimensional barcodes such as Universal Product Code (UPC) barcodes, multi-dimensional barcodes such as Quick Response (QR) codes, Aztec codes, Data Matrix, Dataglyph, MaxiCode, PDF417, Ultra Code, UCC RSS-2D barcodes, and other optical codes), or an acoustic detection component (e.g., a microphone for identifying tagged audio signals). Additionally, various information can be obtained via the communication component 1036, such as location via Internet Protocol (IP) geolocation, location via Wi-Fi® signal triangulation, location via detecting NFC beacon signals that can indicate a specific location.

[0177] Various memories (e.g., main memory 1012, static memory 1014, and the memory of processor 1002) and storage units 1016 may store one or more sets of instructions and data structures (e.g., software) implemented or used by any one or more of the methods or functions described herein. These instructions (e.g., instruction 1008) cause various operations to implement the disclosed examples when executed by processor 1002.

[0178] Instructions 1008 can be sent or received over network 1020 via a network interface device (e.g., a network interface component included in communication component 1036), using a transmission medium and any of a plurality of known transmission protocols (e.g., Hypertext Transfer Protocol (HTTP)). Similarly, instructions 1008 can be sent or received via a transmission medium coupled to device 1022 (e.g., peer-to-peer coupling).

[0179] Software Architecture

[0180] Figure 11This is a block diagram 1100 illustrating a software architecture 1104, which can be installed on any one or more of the devices described herein. The software architecture 1104 is supported by hardware such as a machine 1102 including a processor 1120, memory 1126, and I / O components 1138. In this example, the software architecture 1104 can be conceptualized as a stack of layers, where each layer provides specific functionality. The software architecture 1104 includes layers such as an operating system 1112, libraries 1110, frameworks 1108, and applications 1106. Operationally, application 1106 triggers API calls 1150 through the software stack and receives messages 1152 in response to API calls 1150.

[0181] Operating system 1112 manages hardware resources and provides public services. Operating system 1112 includes, for example, a kernel 1114, services 1116, and drivers 1122. Kernel 1114 acts as an abstraction layer between the hardware layer and other software layers. For example, kernel 1114 provides memory management, processor management (e.g., scheduling), component management, networking and security settings, and other functions. Services 1116 can provide other public services to other software layers. Drivers 1122 are responsible for controlling or interfacing with the underlying hardware. For example, drivers 1122 may include display drivers, camera drivers, BLUETOOTH® or BLUETOOTH® low-power drivers, flash drives, serial communication drivers (e.g., USB drives), Wi-Fi® drivers, audio drivers, power management drivers, etc.

[0182] Library 1110 provides common low-level infrastructure used by application 1106. Library 1110 may include system library 1118 (e.g., the C standard library), which provides functions such as memory allocation, string manipulation, and mathematical functions. Furthermore, library 1110 may include API library 1124, such as media libraries (e.g., libraries for supporting the rendering and manipulation of various media formats, such as Moving Picture Experts Group 4 (MPEG4), Advanced Video Coding (H.264 or AVC), Moving Picture Experts Group Layer 3 (MP3), Advanced Audio Coding (AAC), Adaptive Multi-Rate (AMR) audio codec, Joint Picture Experts Group (JPEG or JPG), or Portable Web Graphics (PNG)), graphics libraries (e.g., the OpenGL framework for rendering graphic content on a display in two-dimensional (2D) and three-dimensional (3D) formats), database libraries (e.g., SQLite, which provides various relational database functions), web libraries (e.g., WebKit, which provides web browsing capabilities), and so on. Library 1110 may also include various other libraries 1128 to provide many other APIs to application 1106.

[0183] Framework 1108 provides common high-level infrastructure for use by application 1106. For example, framework 1108 provides a variety of graphical user interface (GUI) functions, advanced resource management, and advanced location services. Framework 1108 can provide a wide range of other APIs that can be used by application 1106, some of which may be specific to a particular operating system or platform.

[0184] In the example, application 1106 may include home application 1136, contacts application 1130, browser application 1132, book reader application 1134, location application 1142, media application 1144, messaging application 1146, game application 1148, and a variety of other applications such as external application 1140. Application 1106 is a program that performs the functions defined in the program. One or more applications 1106 can be created using various programming languages, such as object-oriented programming languages ​​(e.g., Objective-C, Java, or C++) or procedural programming languages ​​(e.g., C or assembly language). In a particular example, external application 1140 (e.g., an application developed by an entity other than a platform-specific vendor using the Android™ or iOS™ Software Development Kit (SDK)) may be mobile software that runs on a mobile operating system such as iOS™, Android™, Windows® Phone, or another mobile operating system. In this example, an external application 1140 can trigger API calls 1150 provided by the operating system 1112 to facilitate the functions described herein.

[0185] Glossary:

[0186] In this context, "carrier signal" refers to any intangible medium or other intangible medium capable of storing, encoding, or carrying transient or non-transient instructions executed by a machine and including digital or analog communication signals to facilitate the communication of such instructions. Instructions can be sent or received over a network via a network interface device using transient or non-transient transmission media and using any of several well-known transmission protocols.

[0187] In this context, "client device" refers to any machine that interfaces with a communication network to obtain resources from one or more server systems or other client devices. Client devices can be, but are not limited to, mobile phones, desktop computers, laptop computers, PDAs, smartphones, tablet computers, ultrabooks, netbooks, laptop computers, multiprocessor systems, microprocessor-based or programmable consumer electronics, game consoles, set-top boxes, or any other communication device that a user can use to access the network.

[0188] In this context, "communication network" refers to one or more parts of a network, which can be an ad hoc network, intranet, extranet, virtual private network (VPN), local area network (LAN), wireless LAN (WLAN), wide area network (WAN), wireless WAN (WWAN), metropolitan area network (MAN), the Internet, a part of the Internet, a part of the Public Switched Telephone Network (PSTN), a Common Old-Style Telephone Service (POTS) network, a cellular telephone network, a wireless network, a Wi-Fi® network, another type of network, or a combination of two or more such networks. For example, a network or part of a network may include a wireless network or a cellular network, and the coupling may be a Code Division Multiple Access (CDMA) connection, a Global System for Mobile Communications (GSM) connection, or other types of cellular or wireless coupling. In this example, coupling can implement any data transmission technology of various types, such as single-carrier radio transmission technology (1xRTT), evolved data optimization (EVDO) technology, general packet radio service (GPRS) technology, enhanced data rate evolution of GSM (EDGE) technology, the 3rd Generation Partnership Project (3GPP) including 3G, fourth-generation wireless (4G) networks, Universal Mobile Telecommunications System (UMTS), High-Speed ​​Packet Access (HSPA), Global Microwave Access Interoperability (WiMAX), Long Term Evolution (LTE) standards, other data transmission technologies defined by various standards setting organizations, other long-distance protocols, or other data transmission technologies.

[0189] In this context, a "brief message" refers to a message that is accessible for a limited time. Brief messages can be text, images, videos, etc. The access time for a brief message can be set by the message sender. Alternatively, the access time can be a default setting or a setting specified by the recipient. Regardless of the setting method, the message is temporary.

[0190] In this context, "machine-readable medium" means a component, device, or other tangible medium capable of temporarily or permanently storing instructions and data, and may include, but is not limited to, random access memory (RAM), read-only memory (ROM), buffer memory, flash memory, optical media, magnetic media, cache memory, other types of storage devices (e.g., erasable programmable read-only memory (EEPROM)), and / or any suitable combination thereof. The term "machine-readable medium" should be understood to include a single medium or multiple media capable of storing instructions (e.g., a centralized or distributed database, or associated caches and servers). The term "machine-readable medium" should also be considered to include any medium or combination of media capable of storing machine-executable instructions (e.g., code) such that, when executed by one or more processors of the machine, the instructions cause the machine to perform any or more of the methods described herein. Therefore, "machine-readable medium" refers to a single storage device or apparatus, as well as a "cloud-based" storage system or storage network comprising multiple storage devices or apparatuses. The term "machine-readable medium" does not include the signal itself.

[0191] In this context, a “component” refers to a device, physical entity, or logic having boundaries defined by function or subroutine calls, branch points, APIs, or other technologies that provide for partitioning or modularizing specific processing or control functions. A component can be combined with other components via its interface to perform machine processes. A component can be an encapsulated functional hardware unit designed for use with other components and can be part of a program that typically performs a specific function within a related function. A component can constitute a software component (e.g., code implemented on a machine-readable medium) or a hardware component. A “hardware component” is a tangible unit capable of performing certain operations and can be configured or arranged in some physical manner. In various examples, one or more computer systems (e.g., standalone computer systems, client computer systems, or server computer systems) or one or more hardware components (e.g., processors or processor groups) of a computer system can be configured by software (e.g., an application or application portion) to operate to perform certain operations as described herein.

[0192] Hardware components can also be implemented mechanically, electronically, or in any suitable combination thereof. For example, a hardware component may include a dedicated circuit system or logic permanently configured to perform certain operations. A hardware component may be a dedicated processor, such as a field-programmable gate array (FPGA) or an ASIC. A hardware component may also include programmable logic or a circuit system temporarily configured by software to perform certain operations. For example, a hardware component may include software executed by a general-purpose processor or other programmable processor. Once configured by such software, the hardware component becomes a specific machine (or a specific part of a machine) uniquely tailored to perform the configured function and is no longer a general-purpose processor. It will be understood that the decision to implement a hardware component mechanically in a dedicated and permanently configured circuit system or in a temporarily configured (e.g., software-configured) circuit system may be made for cost and time considerations. Accordingly, the phrase “hardware component” (or “hardware-implemented component”) should be understood to include tangible entities, i.e., entities physically constructed, permanently configured (e.g., hardwired), or temporarily configured (e.g., programmed) to operate or perform certain operations described herein. Consider an example where hardware components are temporarily configured (e.g., programmed), and each of these components does not need to be configured or instantiated at any given time. For example, in cases where the hardware components include a general-purpose processor that is configured by software to become a dedicated processor, the general-purpose processor can be configured at different times as its own distinct dedicated processor (e.g., comprising different hardware components). Thus, software configures one or more specific processors to constitute a particular hardware component at one time and different hardware components at different times.

[0193] Hardware components can provide information to and receive information from other hardware components. Therefore, the described hardware components can be considered communicatively coupled. In the presence of multiple hardware components, communication can be achieved (e.g., via appropriate circuitry and buses) through signal transmission between or among two or more hardware components. In examples where multiple hardware components are configured or instantiated at different times, such communication between hardware components can be achieved, for example, by storing information in a memory structure accessible to the multiple hardware components and retrieving information from said memory structure. For example, a hardware component can perform an operation and store the output of that operation in a memory device communicatively coupled to it. Other hardware components can then access the memory device at a subsequent time to retrieve the stored output and process it.

[0194] Hardware components can also initiate communication with input or output devices and operate on resources (e.g., collections of information). Various operations of the example methods described herein can be performed at least partially by one or more processors, which may be temporarily configured (e.g., by software) or permanently configured to perform the relevant operations. Whether the processors are temporarily or permanently configured, such processors can constitute processor-implemented components that operate to perform one or more operations or functions described herein. As used herein, "processor-implemented component" refers to a hardware component implemented using one or more processors. Similarly, the methods described herein can be implemented at least partially by processors, where a particular processor or one or more processors are examples of hardware. For example, at least some of the operations of the methods can be performed by one or more processors or processor-implemented components. Furthermore, one or more processors can also operate to support the execution of relevant operations in a "cloud computing" environment or as "Software as a Service" (SaaS) operations. For example, at least some of the operations can be performed by a group of computers (as an example of a machine including processors), where these operations can be accessed via a network (e.g., the Internet) and via one or more appropriate interfaces (e.g., APIs). The execution of certain operations in an operation can be distributed across processors, residing not only on a single machine but also deployed across multiple machines. In some examples, the processor or processor-implemented components may reside in a single geographic location (e.g., within a home environment, office environment, or server cluster). In other examples, the processor or processor-implemented components may be distributed across several geographic locations.

[0195] In this context, a "processor" refers to any one or more circuits or virtual circuits (e.g., physical circuits simulated by logic executed on an actual processor) that manipulate data values ​​according to control signals (e.g., commands, opcodes, machine codes, control words, macro instructions, etc.) and generate corresponding output signals for operating a machine. A processor may include, for example, at least one of the following: a central processing unit (CPU), a reduced instruction set computing (RISC) processor, a complex instruction set computing (CISC) processor, a graphics processing unit (GPU), a digital signal processor (DSP), a tensor processing unit (TPU), a neural processing unit (NPU), a vision processing unit (VPU), a machine learning accelerator, an artificial intelligence accelerator, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a radio frequency integrated circuit (RFIC), a neuromorphic processor, a quantum processor, or any combination thereof.

[0196] The processor can also be a multi-core processor, which has two or more independent processors (sometimes called "cores") capable of executing instructions simultaneously. A multi-core processor contains multiple computational cores on a single integrated circuit die, each of which can independently execute program instructions in parallel. Parallel processing on a multi-core processor can be implemented via architectures such as superscalar, VLIW, vector processing, or SIMD, which enable each core to run a separate stream of instructions simultaneously.

[0197] A processor can be emulated as a virtual processor or virtual circuit in software running on a physical processor. A virtual processor can behave like a standalone processor, but is implemented in software rather than hardware.

[0198] Changes and modifications may be made to the disclosed examples without departing from the scope of this disclosure. Such and other changes or modifications are intended to be included within the scope of this disclosure as set forth in the appended claims.

[0199] Modules, components and logic

[0200] Some examples are described herein as including logic or multiple components, modules, or mechanisms. Modules can constitute software modules (e.g., code implemented on a machine-readable medium or in transmitted signals) or hardware modules. A “hardware module” is a tangible unit capable of performing a particular operation and can be configured or arranged in a particular physical manner. In various examples, one or more computer systems (e.g., standalone computer systems, client computer systems, or server computer systems) or one or more hardware modules (e.g., processors or processor groups) of a computer system are configured by software (e.g., applications or application portions) to operate to perform some of the operations described herein.

[0201] In some examples, hardware modules are implemented mechanically, electronically, or in any suitable combination thereof. For example, a hardware module may include a dedicated circuit system or logic permanently configured to perform certain operations. For example, a hardware module may be a dedicated processor, such as a field-programmable gate array (FPGA) or an application-specific integrated circuit (ASIC). A hardware module may also include programmable logic or circuit systems temporarily configured by software to perform certain operations. For example, a hardware module may include software contained within a general-purpose processor or other programmable processor. It should be understood that the decision to implement a hardware module mechanically, in a dedicated and permanently configured circuit system, or in a temporarily configured circuit system (e.g., a software-configured circuit system) may be driven by cost and time considerations.

[0202] Therefore, the phrase "hardware module" should be understood to include tangible entities, namely, entities that are physically constructed, permanently configured (e.g., hardwired), or temporarily configured (e.g., programmed) to operate or perform certain operations described herein. As used herein, "hardware-implemented module" refers to a hardware module. Considering examples where hardware modules are temporarily configured (e.g., programmed), each of the hardware modules need not be configured or instantiated at any given time. For example, in cases where the hardware module includes a general-purpose processor configured by software as a dedicated processor, the general-purpose processor can be configured at different times as its own distinct dedicated processors (e.g., including different hardware modules). Thus, software can configure one or more specific processors to constitute a particular hardware module at one time and different hardware modules at different times.

[0203] Hardware modules can provide information to other hardware modules and receive information from other hardware modules. Therefore, the described hardware modules can be considered communication-coupled. When multiple hardware modules exist simultaneously, communication can be achieved through signal transmission between or among two or more hardware modules (e.g., via appropriate circuitry and buses). In examples where multiple hardware modules are configured or instantiated at different times, such communication between or among the hardware modules can be achieved, for example, by storing and retrieving information in a memory structure accessible to the multiple hardware modules. For example, one hardware module performs an operation and stores the output of that operation in a communication-coupled memory device. Another hardware module can then access the memory device at a later time to retrieve and process the stored output. Hardware modules can also initiate communication with input or output devices and can operate on resources (e.g., collections of information).

[0204] The various operations of the example methods described herein can be performed at least in part by one or more processors, which are either temporarily configured (e.g., by software) or permanently configured to perform the relevant operations. Whether temporarily or permanently configured, such processors constitute processor-implemented modules that operate to perform one or more operations or functions described herein. As used herein, "processor-implemented module" refers to a hardware module implemented using one or more processors.

[0205] Similarly, the methods described herein can be implemented at least in part by a processor, where one or more specific processors are examples of hardware. For example, at least some of the operations of the methods can be executed by one or more processors or modules implemented by processors. Furthermore, one or more processors can also operate to support the execution of related operations in a “cloud computing” environment or as “Software as a Service” (SaaS). For example, at least some of the operations can be executed by a group of computers (as an example of a machine including processors), where these operations can be accessed via a network (e.g., the Internet) and via one or more appropriate interfaces (e.g., APIs).

[0206] The execution of certain operations in an operation can be distributed across processors, residing not only on a single machine but also deployed across multiple machines. In some examples, the processor or the module implemented by the processor resides in a single geographic location (e.g., within a home environment, office environment, or server cluster). In other examples, the processor or the module implemented by the processor is distributed across several geographic locations.

Claims

1. A method comprising: Access the uncorrected gaze vector calculated based on the center point of the user's pupil; The image of the pupil is processed by a machine learning model to predict the estimation error in the gaze vector. The machine learning model is trained to establish a relationship between multiple ground-value gaze vectors and uncorrected gaze vectors for multiple pupil parameters, including diameter, gaze angle, or gaze eccentricity. as well as A corrected gaze vector is generated by applying the estimation error in the gaze vector predicted by the machine learning model to the uncorrected gaze vector, which has been calculated based on the center point of the user's pupil.

2. The method according to claim 1, further comprising: One or more augmented reality operations are performed based on the corrected gaze vector.

3. The method according to any one of claims 1 to 2, further comprising: Identify the object depicted in the display of the augmented reality device that corresponds to the corrected gaze vector; as well as Perform one or more augmented reality operations related to the identified object.

4. The method according to any one of claims 1 to 3, wherein, The machine learning model is also trained to establish a relationship between the plurality of ground truth gaze vectors and the uncorrected gaze vectors based on gaze eccentricity.

5. The method according to any one of claims 1 to 4, further comprising: The first component of the machine learning model is trained to predict a first estimation error in the first angular component; The second component of the machine learning model is trained to predict a second estimation error in the second angular component; as well as The third component of the machine learning model is trained to predict the third estimation error in the third angular component.

6. The method according to claim 5, wherein, The first component, the second component, and the third component are trained individually and independently of each other.

7. The method according to any one of claims 1 to 6, wherein, The first, second, and third components are trained for each of the user's multiple eyes to enable the correction to be applied to a combination of uncorrected gaze vectors associated with the user's left and right eyes, or the correction to be applied individually to a first uncorrected gaze vector associated with the left eye and a second uncorrected gaze vector associated with the right eye.

8. The method according to any one of claims 1 to 7, further comprising: Access multiple training data sets, including the uncorrected gaze vectors for the multiple pupil diameters and the multiple ground truth gaze vectors associated with the uncorrected gaze vectors for the multiple pupil diameters; Obtain the first batch of training data, which includes a first uncorrected gaze vector for a first pupil diameter; The machine learning model is used to process the first pupil diameter to predict a first estimation error in the first uncorrected gaze vector; Calculate the truth error between the first uncorrected gaze vector and the first true gaze vector associated with the first uncorrected gaze vector; Calculate the deviation between the true error and the first estimation error; as well as The calculated bias is used to update one or more parameters of the machine learning model.

9. The method according to claim 8, wherein, The first estimation error includes multiple estimation errors for different angular components, and the method further includes: Calculate multiple truth errors, each truth error being associated with a different angular component of the first uncorrected gaze vector and the first true gaze vector; and For different angular components, calculate multiple deviations between the multiple true errors and the multiple estimated errors.

10. The method according to any one of claims 1 to 9, wherein, The machine learning model is updated based on the aforementioned biases, including one or more parameters.

11. The method according to any one of claims 1 to 10, wherein, The truth error is calculated based on the first pupil diameter, the square of the first pupil diameter, and the reported fixation eccentricity associated with the first pupil diameter.

12. The method according to any one of claims 1 to 11, further comprising generating at least a portion of the plurality of training data by performing a calibration operation, the calibration operation comprising: A stimulus is presented on a display, the stimulus having a known location; Determine the training pupil diameter; The uncorrected gaze vector is calculated based on the detected pupil diameter; The known gaze vector is calculated based on the correlation between the detected pupil diameter and the known location of the stimulus; as well as The trained uncorrected gaze vector and the detected pupil diameter associated with the known gaze vector are stored as a portion of the plurality of training data.

13. The method of claim 12, further comprising: Repeat the calibration operation for multiple background light intensities.

14. The method according to any one of claims 1 to 13, wherein, The multiple background light intensities range from 0.5 nits to 2000 nits to capture the full range of pupil diameters expected in an unconstrained setting.

15. The method according to any one of claims 1 to 14, wherein, The uncorrected gaze vector is calculated based on corneal reflection and the center position of the pupil.

16. The method according to any one of claims 1 to 15, further comprising detecting changes in the detection point of the pupil based on different environmental conditions, the changes in the detection point causing an error in the calculated uncorrected gaze vector, the error including at least one of errors in gaze position, direction, or convergence depth estimation.

17. The method of any one of claims 1 to 16, further comprising adding the estimated error in the gaze vector to the uncorrected gaze vector to generate the corrected gaze vector.

18. A system comprising: At least one storage device; as well as At least one processor, coupled to the at least one storage device and configured to perform operations including: Access the uncorrected gaze vector calculated based on the center point of the user's pupil; The pupil image is processed using a machine learning model to predict estimation errors in the gaze vector. The machine learning model is trained to establish relationships between multiple ground-value gaze vectors and uncorrected gaze vectors for multiple pupil parameters, including diameter, gaze angle, or gaze eccentricity. A corrected gaze vector is generated by applying the estimation error in the gaze vector predicted by the machine learning model to the uncorrected gaze vector, which has been calculated based on the center point of the user's pupil.

19. The system according to claim 18, wherein, The operation includes: One or more augmented reality operations are performed based on the corrected gaze vector.

20. A non-transitory machine-readable storage medium comprising instructions that, when executed by one or more processors of a machine, cause the machine to perform operations, the operations including: Access the uncorrected gaze vector calculated based on the center point of the user's pupil; The image of the pupil is processed by a machine learning model to predict the estimation error in the gaze vector. The machine learning model is trained to establish a relationship between multiple ground-value gaze vectors and uncorrected gaze vectors for multiple pupil parameters, including diameter, gaze angle, or gaze eccentricity. as well as A corrected gaze vector is generated by applying the estimation error in the gaze vector predicted by the machine learning model to the uncorrected gaze vector, which has been calculated based on the center point of the user's pupil.

Images