Body mesh reconstruction from RGB images

Abstract

Methods and systems for generating a body mesh from a single image are disclosed. The system predicts both a volumetric reconstruction tensor of a monocular image and a pose of an object by applying a first machine learning model to the monocular image. The system identifies a portion of a pose of an object corresponding to a point in a canonical space associated with the set of location encoding information. The system obtains points of a volumetric reconstruction tensor corresponding to the identified portion of the pose. The system classifies the obtained points as being inside or outside the canonical volume by applying a second machine learning model together with the set of position encoding information to the points of the obtained volume reconstruction tensor. The system generates a three-dimensional (3D) mesh representing an object in a normative space.

Description

[0001] Priority requirements

[0002] This application claims the benefit of priority to Greek patent application serial number 20230100328, filed April 13, 2023, and U.S. patent application serial number 18 / 339,780, filed June 22, 2023, each of which is incorporated herein by reference in its entirety. Technical Field

[0003] This disclosure relates to extended reality (XR) devices, such as augmented reality (AR) devices and / or virtual reality (VR) devices. Background Technology

[0004] AR is a modification of the virtual environment. For example, in VR, the user is fully immersed in the virtual world, while in AR, the user is immersed in a world where virtual objects are combined or overlaid on the real world. AR systems are designed to generate and present virtual objects that interact realistically with the real-world environment and with each other. Examples of AR applications can include single-player or multiplayer video games, instant messaging systems, and more. Attached Figure Description

[0005] In the accompanying drawings (which are not necessarily drawn to scale), similar reference numerals may describe similar parts in different views. To facilitate identification of any particular element or action being discussed, one or more of the highest-order digits in the reference numerals indicate the drawing number in which the element was first introduced. Some non-limiting examples are shown in the accompanying drawings:

[0006] Figure 1 It is a graphical representation of a networked environment in which the present disclosure can be deployed, based on some examples.

[0007] Figure 2 It is a graphical representation of a messaging system with both client-side and server-side functionalities, based on some examples.

[0008] Figure 3 It is a graphical representation of the data structures maintained in the database based on some examples.

[0009] Figure 4 It is a graphical representation based on some example messages.

[0010] Figure 5 It is a graphical representation of a mesh generation system based on some examples.

[0011] Figure 6 It is a graphical representation of a whole-body normative space reconstruction system based on some examples.

[0012] Figure 7 It is a graphical representation of the training loss of a whole-body canonical space reconstruction system based on some examples.

[0013] Figure 8 It is a detailed graphical representation of head pose based on some examples of a full-body canonical space reconstruction system.

[0014] Figure 9 It is a graphical representation of the mesh generation components of a whole-body canonical space reconstruction system based on some examples.

[0015] Figure 10 This is a flowchart illustrating example operations of a mesh generation system based on some examples.

[0016] Figure 11 It is a graphical representation of a machine in the form of a computer system, based on some examples, within which a set of instructions can be executed to cause the machine to perform any or more of the methods discussed herein.

[0017] Figure 12 It is a block diagram showing the software architecture in which examples can be implemented.

[0018] Figure 13 The system shown is one of several examples in which a head-mounted wearable device can be implemented. Detailed Implementation

[0019] The following description includes systems, methods, techniques, instruction sequences, and computer program products embodying illustrative examples of this disclosure. In this description, numerous specific details are set forth for illustrative purposes in order to provide an understanding of the various examples. However, it will be apparent to those skilled in the art that the examples described can be practiced without these specific details. Generally, well-known examples of instructions, protocols, structures, and techniques need not be shown in detail.

[0020] Typically, VR and AR systems display an image representing a given user by capturing their image, and additionally, obtain a depth map using a depth sensor that depicts a real-world human body within the image. By processing the depth map and the image together, VR and AR systems can detect the user's location within the image and can appropriately modify the user or background within the image. While such systems function well, the requirement for depth sensors limits their applications. This is because adding a depth sensor to the user's device for image modification increases the overall cost and complexity of the device, making them less attractive and more difficult to implement in mobile device settings.

[0021] Some systems estimate the three-dimensional (3D) mesh of a person depicted in an image to provide AR / VR experiences. Understanding human 3D pose is crucial for human-computer interaction, motion capture, action recognition, and augmented reality and virtual reality (AR / VR). Recently, human mesh reconstruction has received increasing attention because it unlocks the prerequisites for new experiences requiring detailed estimation of 3D human geometry, such as avatar construction and telepresence in VR or visual effects and virtual try-on in AR, based on information beyond the 3D skeleton.

[0022] Despite rapid advancements in 3D mesh reconstruction techniques, meshes generated by current systems remain far from pixel-accurate when projected back into the image domain. Some conventional systems use cropped images to predict 3D meshes using parameterized human models. Convolutional encoders minimize reprojection errors at 2D joints, while discriminators determine the reasonableness of 3D parameters. Such systems are highly complex to implement (e.g., on mobile platforms) and still produce results lacking pixel accuracy. Some conventional systems generate 3D meshes by using multiple networks to predict full-body 3D reconstructions of objects depicted in an image, each trained independently of the others. For example, a conventional system might use a first machine learning model to generate the pose of the object depicted in the image, a second machine learning model to generate pixel representations of the object, and a third machine learning model to determine whether the object's pixels correspond to the canonical spatial volume representing the object. Because these conventional systems use three separate machine learning models to generate independent predictions, the results can include inconsistencies, poor training, and inefficiencies when deployed on user devices such as mobile platforms. Even so, these systems cannot provide pixel-accurate representations of both the user's entire body and facial features, and often produce many blurry mesh areas, particularly in the facial region.

[0023] The disclosed technique attempts to improve the way 3D meshes are generated for one or more real-world objects (e.g., real-world people) depicted in an image or video. The disclosed technique predicts both the volumetric reconstruction tensor of the monocular image and the pose of the object by applying a first machine learning model to the monocular image depicting the object. The disclosed technique identifies portions of the object's pose corresponding to points in a canonical space associated with a set of position-encoded information. The disclosed technique obtains points in the volumetric reconstruction tensor corresponding to the identified portions of the pose. The disclosed technique classifies the obtained points as either inside or outside the canonical volume by applying a second machine learning model, along with the set of position-encoded information, to the points in the obtained volumetric reconstruction tensor. The disclosed technique generates a 3D mesh representing the object in the canonical space.

[0024] Furthermore, the disclosed technique applies similar processing to generate pixel-accurate facial meshes of objects depicted in an image. Specifically, the disclosed technique crops a portion of a monocular image corresponding to the object's head and predicts a head-specific canonical spatial reconstruction of the object by applying a head-specific canonical network based on the generated facial landmarks of the object to the cropped portion. The disclosed technique then combines the head-specific canonical spatial reconstruction with the canonical volume corresponding to the object to generate a pixel-accurate canonical volume reconstruction of the object. In some cases, the disclosed technique smoothly deforms the overlapping regions between the full-body canonical spatial reconstruction of the object and the head-specific canonical spatial reconstruction to generate a pixel-accurate canonical volume reconstruction. This pixel-accurate canonical volume reconstruction can be used to generate and provide a bottom-up estimate of the 3D mesh geometry, which can be projected back to the image in a pixel-accurate manner.

[0025] This simplifies the process of adding AR graphics to images or videos, significantly reducing design constraints and costs when generating such AR graphics, and decreasing processing complexity, power consumption, and memory requirements. It also improves the illusion that AR graphics are part of a real-world environment depicted in an image or video representing a real-world object. This enables the seamless and efficient addition of AR graphics to underlying images or videos in real time on small mobile devices. The disclosed technology can be applied specifically or primarily to mobile devices without requiring the mobile device to send images / videos to a server. In other examples, the disclosed technology is applied specifically or primarily to a remote server, or it can be partitioned between the mobile device and the server.

[0026] This improves the overall user experience when using electronic devices. Furthermore, by providing such an AR experience without using a depth sensor, the total amount of system resources required to complete the task is reduced. As used herein, “clothing,” “fashion items,” and “outfits” are used interchangeably and should be understood to have the same meaning. Clothing, apparel, or fashion items can include shirts, skirts, dresses, shoes, wallets, furniture items, household items, glasses, eyeglasses, AR logos, AR badges, trousers, shorts, jackets, t-shirts, blouses, glasses, jewelry, earrings, earmuffs, hats, earmuffs, or any other suitable item or object.

[0027] Networked computing environment

[0028] Figure 1This is a block diagram illustrating an example interactive system 100 for facilitating interactions on a network, such as exchanging text messages, making text audio and video calls, or playing games. Interactive system 100 includes multiple user systems 102, each of which hosts multiple applications including an interactive client 104 and other applications 106. Each interactive client 104 is communicatively coupled to other instances of the interactive client 104 (e.g., hosted on corresponding other user systems 102), an interactive server system 110, and a third-party server 112 via one or more communication networks including a network 108 (e.g., the Internet). The interactive client 104 may also communicate with the locally hosted applications 106 using an application programming interface (API).

[0029] Each user system 102 may include multiple user devices, such as mobile device 114, head-mounted wearable device 116, and computer client device 118, which can be communicatively connected to exchange data and messages.

[0030] Interactive client 104 interacts with other interactive clients 104 and with interactive server system 110 via network 108. The data exchanged between interactive clients 104 (e.g., interaction 120) and between interactive client 104 and interactive server system 110 includes functions (e.g., commands for activating functions) and payload data (e.g., text, audio, video, or other multimedia data).

[0031] Interactive server system 110 provides server-side functionality to interactive client 104 via network 108. While some functions of interactive system 100 are described herein as being performed by interactive client 104 or interactive server system 110, the location of certain functions within interactive client 104 or interactive server system 110 may be a design choice. For example, it may be technically preferred that specific technologies and functions are initially deployed within interactive server system 110, but later migrated to interactive client 104 of user system 102 with sufficient processing power.

[0032] The interactive server system 110 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 message content, client device information, geolocation information, media enhancements and overlays, message content persistence conditions, entity relationship information, and live event information. Data exchange within the interactive system 100 is activated and controlled via functions available through the user interface of the interactive client 104.

[0033] Now, specifically, the interaction server system 110 is used. API server 122 is coupled to interaction server 124 and provides it with a programming interface, making the functionality of interaction server 124 accessible to interaction client 104, other applications 106, and third-party server 112. Interaction server 124 is communicatively coupled to database server 126, thereby facilitating access to database 128, which stores data associated with the interactions processed by interaction server 124. Similarly, web server 130 is coupled to interaction server 124 and provides a web-based interface to interaction server 124. For this purpose, web server 130 handles incoming network requests via Hypertext Transfer Protocol (HTTP) and several other related protocols.

[0034] API server 122 receives and sends interactive data (e.g., command and message payloads) between interactive server 124 and user system 102 (as well as interactive client 104 and other applications 106) and third-party server 112. Specifically, API server 122 provides a set of interfaces (e.g., routines and protocols) that interactive client 104 and other applications 106 can call or query to activate the functionality of interactive server 124. API server 122 exposes various functions supported by interactive server 124, including account registration; login functionality; sending interactive data from one interactive client 104 to another interactive client 104 via interactive server 124; transferring media files (e.g., images or videos) from interactive client 104 to interactive server 124; setting media data sets (e.g., stories); retrieving the friend list of users in user system 102; retrieving messages and content; adding and deleting entities (e.g., friends) against an entity graph (e.g., entity graph 310); locating friends within the entity graph; and opening (e.g., application events associated with interactive client 104).

[0035] Interactive server 124 hosts multiple systems and subsystems, as shown below. Figure 2 Describe it.

[0036] Application of links

[0037] Returning to interactive client 104, the features and functionality of external resources (e.g., linked application 106 or applet) are available to the user via the interface of interactive client 104. In this context, "external" refers to the fact that application 106 or applet is outside of interactive client 104. External resources are typically provided by third parties, but may also be provided by the creator or provider of interactive client 104. Interactive client 104 receives user selections regarding options for launching or accessing the features of such external resources. External resources may be application 106 installed on user system 102 (e.g., a "local app"), or a smaller version (e.g., a "app") of an application hosted on user system 102 or located remotely on user system 102 (e.g., on a third-party server 112). A smaller version of an application includes a subset of the features and functionality of the application (e.g., a full-scale local version of the application) and is implemented using markup language documentation. In some examples, a smaller version of an application (e.g., an "app") is a web-based markup language version of the application and is embedded in interactive client 104. In addition to using markup language documentation (e.g., ...), other features may also be implemented using markup language documentation. In addition to files, mini-programs can include scripting languages ​​(e.g., ...). (files or .json files) and stylesheets (e.g., ...). document).

[0038] In response to receiving a user selection of an option for launching or accessing an external resource, interactive client 104 determines whether the selected external resource is a web-based external resource or a locally installed application 106. In some cases, application 106, locally installed on user system 102, can be launched independently of and separately from interactive client 104, for example, by selecting the icon corresponding to application 106 on the home screen of user system 102. A smaller version of such an application can be launched or accessed via interactive client 104, and in some examples, no part of the smaller application can be accessed outside of interactive client 104, or only a limited portion of the smaller application can be accessed outside of interactive client 104. A smaller application can be launched by receiving, for example, markup language documents associated with the smaller application from third-party server 112 and processing such documents via interactive client 104.

[0039] In response to determining that the external resource is a locally installed application 106, the interactive client 104 instructs the user system 102 to launch the external resource by executing locally stored code corresponding to the external resource. In response to determining that the external resource is a web-based resource, the interactive client 104 communicates with a third-party server 112 (e.g.) to obtain a markup language document corresponding to the selected external resource. The interactive client 104 then processes the obtained markup language document to present the web-based external resource within the user interface of the interactive client 104.

[0040] Interactive client 104 can notify users of user system 102 or other users (e.g., "friends") associated with such users of one or more external resources of ongoing activity. For example, interactive client 104 can provide participants in a conversation (e.g., a chat conversation) within interactive client 104 with notifications related to external resources currently or recently used by one or more members of a user group. One or more users can be invited to join an active external resource or to activate a recently used but currently inactive external resource (in a friend group). External resources can provide participants in the conversation, each using their respective interactive client 104, with the ability to share items, conditions, states, or locations within the external resource with one or more members of the user group during the chat session. Shared items can be interactive chat cards that chat members can use to interact, for example, activate the corresponding external resource, view specific information within the external resource, or take chat members to a specific location or state within the external resource. Within a given external resource, response messages can be sent to users on interactive client 104. External resources can selectively include different media items in the response based on the current context of the external resource.

[0041] Interactive client 104 can present a list of available external resources (e.g., application 106 or mini-program) to the user to launch or access a given external resource. This list can be presented as a context-sensitive menu. For example, the icons representing different applications (or mini-programs) of application 106 (or mini-program) can change based on how the user launches the menu (e.g., from a conversational interface or from a non-conversational interface).

[0042] System Architecture

[0043] Figure 2This is a block diagram illustrating further details of the interactive system 100 based on some examples. Specifically, the interactive system 100 is shown as including an interactive client 104 and an interactive server 124. The interactive system 100 includes multiple subsystems supported on the client side by the interactive client 104 and on the server side by the interactive server 124. Example subsystems are discussed below and may include a mesh generation system 500 that generates a mesh of objects (e.g., the user) depicted in a two-dimensional (2D) image to enable the user to control the interactive client / application / XR experience. The following is in conjunction with... Figure 5 An illustrative implementation of the mesh generation system 500 is shown and described.

[0044] In some examples, these subsystems are implemented as microservices. A microservice subsystem (e.g., a microservice application) can have components that enable it to operate independently and communicate with other services. Example components of a microservice subsystem may include:

[0045] • Functional Logic: Functional logic implements the functions of the microservice subsystem and represents the specific capabilities or functions provided by the microservice.

[0046] • API Interface: Microservices can communicate with other components using lightweight protocols (such as REST or messaging) through well-defined APIs or interfaces. API interfaces define the inputs and outputs of a microservice subsystem, and how it interacts with other microservice subsystems of the interactive system 100.

[0047] • Data storage: The microservice subsystem can be responsible for its own data storage, which can be in the form of a database, cache, or other storage mechanisms (e.g., using database server 126 and database 128). This allows the microservice subsystem to operate independently of other microservices in the interactive system 100.

[0048] • Service Discovery: Microservice subsystems can find and communicate with other microservice subsystems in the interacting system 100. The service discovery mechanism enables microservice subsystems to locate and communicate with other microservice subsystems in a scalable and efficient manner.

[0049] • Monitoring and logging: Microservice subsystems may need to be monitored and logged to ensure availability and performance. Monitoring and logging mechanisms enable the tracking of the health and performance of microservice subsystems.

[0050] In some examples, the interactive system 100 may employ a monolithic architecture, a service-oriented architecture (SOA), a function-as-a-service (FaaS) architecture, or a modular architecture:

[0051] The image processing system 202 provides various functions that enable users to capture and enhance (e.g., annotate or otherwise modify or edit) media content associated with a message.

[0052] The camera device system 204 includes (e.g., in a camera device application) control software that interacts with and controls the camera device hardware of the user system 102 (e.g., directly or via operating system controls) to modify and enhance real-time images captured and displayed via the interactive client 104.

[0053] Enhancement system 206 provides functionality related to the generation and distribution of enhancements (e.g., media overlays) for images captured in real time by the camera device of user system 102 or retrieved from the memory of user system 102. For example, enhancement system 206 is operable to select, present, and display media overlays (e.g., image filters or image lenses) for interactive client 104 to enhance real-time images received via camera device system 204 or retrieved from memory 1302 of user system 102. Figure 13 (As shown in the image) retrieved stored images. These enhancements are selected by enhancement system 206 based on some input and data and presented to the user of interactive client 104, such as:

[0054] • The geographic location of user system 102; and

[0055] • Entity relationship information of users in user system 102.

[0056] Enhancements 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., photos or videos) at user system 102 for transmission in messages, or to video content such as video content streams or feeds sent from interactive client 104. Therefore, image processing system 202 can interact with and support various subsystems of communication system 208, such as messaging system 210 and video communication system 212.

[0057] Media overlays may include text or image data that can be superimposed on photographs taken by user system 102 or video streams produced by user system 102. In some examples, media overlays may be location overlays (e.g., Venice Beach), names of live events, or names of businesses (e.g., beach cafes). In other examples, image processing system 202 uses the geolocation of user system 102 to identify media overlays that include the names of businesses located at the geolocation of user system 102. Media overlays may include additional tags associated with businesses. Media overlays may be stored in database 128 and accessed through database server 126.

[0058] Image processing system 202 provides a user-based publishing platform that allows users to select a geographic location on a map and upload content associated with that location. Users can also specify which media overlays should be provided to other users. Image processing system 202 generates a media overlay that includes the uploaded content and associates it with the selected geographic location.

[0059] The Enhanced Creation System 214 supports AR developer platforms and includes applications that enhance (e.g., AR experiences) interactive clients 104 for content creators (e.g., artists and developers). The Enhanced Creation System 214 provides content creators with a library of built-in features and tools, including, for example, custom shaders, tracking technologies, and templates.

[0060] In some examples, enhancement creation system 214 provides a merchant-based publishing platform that enables merchants to select specific enhancements associated with geolocation via a bidding process. For example, enhancement creation system 214 associates the media overlay of the highest bidder with a corresponding geolocation for a predefined amount of time.

[0061] Communication system 208 is responsible for enabling and processing various forms of communication and interaction within interactive system 100, and includes messaging system 210, audio communication system 216, and video communication system 212. Messaging system 210 is responsible for enabling temporary or time-limited access to content by interactive client 104. Messaging system 210 includes (e.g., in user management system 218) multiple timers that selectively enable access (e.g., for presentation and display) of messages and associated content via interactive client 104 based on duration and display parameters associated with a message or set of messages (e.g., a story). Audio communication system 216 enables and supports audio communication (e.g., real-time audio chat) between multiple interactive clients 104. Similarly, video communication system 212 enables and supports video communication (e.g., real-time video chat) between multiple interactive clients 104.

[0062] User management system 218 is responsible for managing user data and profiles during operation, and maintaining entity information about users of interactive system 100 and the relationships between users (e.g., stored in entity table 308, entity diagram 310, and profile data 302).

[0063] The collection management system 220, during operation, is responsible for managing collections or sets 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 libraries" or "event stories." Such collections can be made available for a specified time period (e.g., the duration of the event to which the content relates). For example, content related to a concert can be available as a "story" for the duration of the concert. The collection management system 220 can also be responsible for publishing icons that notify the user interface of the interactive client 104 of the availability of specific collections. The collection management system 220 includes curation functions that enable collection managers to manage and curate specific content collections. For example, a curation interface enables event organizers to curate collections of content related to a specific event (e.g., to remove inappropriate content or redundant messages). Additionally, the collection management system 220 employs machine vision (or image recognition technology) and content rules to automatically curate content collections. In some examples, users may be compensated for including user-generated content in a collection. In such cases, the collection management system 220 operates to automatically pay such users for using their content.

[0064] Map system 222 provides various geolocation (e.g., location services) functions and supports the presentation of map-based media content and messages by interactive client 104. For example, map system 222 enables the display (e.g., stored on) maps. Figure 3 The user's profile data 302 (in which user icons or avatars are used) indicates the current or past location of the user's "friends" within the context of the map, as well as media content generated by these friends (e.g., a collection of messages including photos and videos). For example, on the map interface of interactive client 104, messages posted by the user from a specific geographic location to interactive system 100 can be displayed to the specific user's "friends" within the context of that specific location on the map. The user can also share his or her 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 within the context of the map interface of interactive client 104.

[0065] Game system 224 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 that can be initiated 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 participate in specific games by sending invitations from interactive client 104. Interactive client 104 also supports sending and receiving voice, video, and text messages (e.g., chat) within the context of playing the game, provides leaderboards for the game, and also supports providing in-game rewards (e.g., game currency and items).

[0066] External resource system 226 provides interactive client 104 with an interface to communicate with remote servers (e.g., third-party server 112) to launch or access external resources (i.e., applications or applets). Each third-party server 112 hosts applications or smaller versions of applications (e.g., game applications, utility applications, payment applications, or ride-sharing applications) based on markup languages ​​(e.g., HTML5). Interactive client 104 can launch web-based resources (e.g., applications) by accessing HTML5 files from the third-party server 112 associated with the web-based resource. The application hosted by third-party server 112 is programmed in JavaScript using a software development kit (SDK) provided by interactive server 124. The SDK includes APIs with functionality that can be called or activated by the web-based application. Interactive server 124 hosts a JavaScript library that provides access to a given external resource for specific user data of interactive client 104. HTML5 is an example of a technology for programming games, but applications and resources programmed using other technologies can be used.

[0067] To integrate the SDK's functionality into the web-based resource, the third-party server 112 downloads the SDK from the interactive server 124, or the third-party server 112 otherwise receives the SDK. 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.

[0068] The SDK stored on the interactive server system 110 effectively bridges the gap between external resources (e.g., application 106 or applet) and the interactive client 104. This provides users with a seamless experience communicating with other users on the interactive client 104 while preserving the appearance of the interactive client 104. To bridge communication between the external resources and the interactive client 104, the SDK facilitates communication between the third-party server 112 and the interactive client 104. A bridging script running on the user system 102 establishes two unidirectional communication channels between the external resources and the interactive client 104. Messages are sent asynchronously between the external resources and the interactive client 104 via these communication channels. Each SDK function activation 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.

[0069] By using the SDK, not all information from the interactive client 104 is shared with the third-party server 112. The SDK limits which information is shared based on the needs of the external resources. Each third-party server 112 provides the interactive server 124 with an HTML5 file corresponding to the web-based external resource. The interactive server 124 can add a visual representation (e.g., box design or other graphics) of the web-based external resource to the interactive client 104. Once the user selects the visual representation or instructs the interactive client 104 to access the features of the web-based external resource through the graphical user interface (GUI), the interactive client 104 obtains the HTML5 file and instantiates the resource for accessing the features of the web-based external resource.

[0070] Interactive client 104 presents a GUI (e.g., a login page or title screen) for an external resource. 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 user data of interactive client 104. In response to determining that the initiated external resource has previously been authorized to access user data of interactive client 104, interactive client 104 presents an additional GUI for the external resource, including its functionality and characteristics. In response to determining that the initiated external resource has not previously been authorized to access 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 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 allows the external resource to access user data from interactive client 104. External resources are authorized by the interactive client 104 to access user data under the OAuth 2 framework.

[0071] Interactive client 104 controls the type of user data shared with external resources based on the type of authorized external resource. For example, it provides access to a first type of user data (e.g., 2D avatars of users with or without different avatar characteristics) to external resources including full-scale applications (e.g., application 106). As another example, it provides access to a second type of user data (e.g., payment information, 2D avatars of users, 3D avatars of users, and avatars with various avatar characteristics) to external resources including smaller versions of applications (e.g., web-based versions of applications). Avatar characteristics include different ways of customizing the appearance of an avatar (e.g., different poses, facial features, clothing, etc.).

[0072] The advertising system 228 enables third parties to purchase advertisements to be presented to end users via the interactive client 104 during operation, and also handles the delivery and presentation of these advertisements.

[0073] Artificial intelligence and machine learning system 230 provides various services to different subsystems within interaction system 100. For example, AI and machine learning system 230 operates in conjunction with image processing system 202 and camera device system 204 to analyze images and extract information such as objects, text, or faces. This information can then be used by image processing system 202 to enhance, filter, or manipulate images. Enhancement system 206 can use AI and machine learning system 230 to generate enhanced content, XR experiences, and AR experiences, such as adding virtual objects or animations to real-world images. Communication system 208 and messaging system 210 can use AI and machine learning system 230 to analyze communication patterns and provide insights into how users interact with each other, and provide intelligent message classification and tagging, such as classifying messages based on sentiment or topic. AI and machine learning system 230 can also provide chatbot functionality to messaging interactions 120 between user systems 102 and between user systems 102 and interaction server system 110. The artificial intelligence and machine learning system 230 can also work with the audio communication system 216 to provide speech recognition and natural language processing capabilities, thereby allowing users to interact with the interactive system 100 using voice commands. In some cases, the artificial intelligence and machine learning system 230 can implement one or more machine learning models that use a combination of head-specific canonical reconstruction networks and / or body canonical reconstruction networks to generate 3D meshes of objects depicted in 2D images.

[0074] Data Architecture

[0075] Figure 3 This is a schematic diagram illustrating a data structure 300 that can be stored in a database 304 of an interactive server system 110, according to certain examples. Although the contents of the database 304 are shown as including multiple tables, it should be understood that data can be stored in other types of data structures, such as object-oriented databases.

[0076] Database 304 includes message data stored in message table 306. 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 provided regarding information that can be included in the message and is contained within the message data stored in message table 306.

[0077] Entity table 308 stores entity data and (for example, links to entity diagram 310 and profile data 302). Entities whose records are maintained in entity table 308 can include individuals, company entities, organizations, objects, locations, events, etc. Regardless of the entity type, any entity whose data is stored by the interactive server system 110 can be an identifiable entity. Each entity is assigned a unique identifier and an entity type identifier (not shown).

[0078] Entity graph 310 stores information about the relationships and associations between entities. As an example only, such relationships can be social, professional (e.g., working in a common company or organization), interest-based, or activity-based. Some relationships between entities can be one-way, such as an individual user subscribing to digital content from a business or publishing user (e.g., a newspaper or other digital media channel or brand). Other relationships can be two-way, such as the "friendship" relationships between the various users of interactive system 100.

[0079] Certain licenses and relationships can be attached to each relationship, and also to each direction of the relationship. For example, a two-way relationship (e.g., a friendship between individual users) can include authorization for the posting of digital content items between the individual users, but certain restrictions or filters can be imposed on the posting of such digital content items (e.g., based on content characteristics, location data, or time of day data). Similarly, a subscription relationship between an individual user and a business user can impose varying degrees of restrictions on the posting of digital content from the business user to the individual user, and can significantly restrict or prevent the posting of digital content from the individual user to the business user. As an example of an entity, a specific user can (e.g., through privacy settings) record certain restrictions in the records for that entity within entity table 308. Such privacy settings can be applied to all types of relationships in the context of interaction system 100, or selectively applied to certain types of relationships.

[0080] Profile data 302 stores various types of profile data about a specific entity. Based on privacy settings specified by the specific entity, profile data 302 can be selectively used and presented to other users of the interactive system 100. In the case of an individual, profile data 302 includes, for example, a username, phone number, address, settings (e.g., notification and privacy settings), and an avatar representation (or a set of such avatar representations) selected by the user. The specific user can then selectively include one or more of these avatar representations within the content of messages transmitted via the interactive system 100 and on a map interface displayed to other users by the interactive client 104. The set of avatar representations may include "status avatars," which present a graphical representation of a status or activity that the user can choose to transmit at a specific time.

[0081] In the case that the entity is a group, in addition to the group name, members and various settings for the relevant group (e.g., notifications), the profile data 302 for the group may similarly include one or more avatars associated with the group.

[0082] Database 304 also stores enhancement data, such as overlays or filters, in enhancement table 312. Enhancement data is associated with and applied to videos (video data is stored in video table 314) and images (image data is stored in image table 316).

[0083] In some examples, filters are overlays displayed as superimposed on images or videos during presentation to the recipient user. Filters can be of various types, including user-selected filters from a set of filters presented to the sending user by the interactive client 104 when the sending user is composing a message. Other types of filters include geolocation filters (also known as geographic filters), which can be presented to the sending user based on geographic location. 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 user system 102.

[0084] 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 user system 102 during the message creation process. Examples of data filters include the current temperature at a specific location, the sending user's current speed, the battery life of the user system 102, or the current time.

[0085] Other augmented data that can be stored within image table 316 includes AR content items (e.g., corresponding to the application "Lens" or AR experience). AR content items can be real-time special effects and sounds that can be added to images or videos.

[0086] Collection table 318 stores data about collections of messages and associated image, video, or audio data, 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 whose records are maintained in entity table 308). 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.

[0087] The collection can also constitute a "live story," which is a collection of content from multiple users created manually, automatically, or using a combination of manual and automatic technologies. For example, a "live story" can constitute a curated stream of user-submitted content from different 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 collective perspective.

[0088] Another type of content collection is called a "location story," which allows users of user system 102 located in a specific geographic location (e.g., on a college or university campus) to contribute to a specific collection. In some examples, contributions to a location story may employ secondary authentication to verify that the end user belongs to a specific organization or other entity (e.g., a student on a university campus).

[0089] As mentioned above, video table 314 stores video data, which in some examples is associated with messages whose records are maintained within message table 306. Similarly, image table 316 stores image data associated with messages whose message data is stored in entity table 308. Entity table 308 can associate various enhancements from enhancement table 312 with various images and videos stored in image table 316 and video table 314.

[0090] Database 304 also includes trained machine learning techniques 307, which store parameters of one or more machine learning models trained during the training of the grid generation system 500. For example, trained machine learning techniques 307 store trained parameters of one or more artificial neural network machine learning models or techniques.

[0091] Data communication architecture

[0092] Figure 4 This is a schematic diagram illustrating the structure of message 400 according to some examples, generated by interactive client 104 for transmission to another interactive client 104 via interactive server 124. The content of a particular message 400 is used to populate message table 306 within database 304 accessible by interactive server 124. Similarly, the content of message 400 is stored in memory as "in-transit" or "in-flight" data for user system 102 or interactive server 124. Message 400 is shown to include the following example components:

[0093] • Message Identifier 402: A unique identifier that identifies message 400.

[0094] • Message text payload 404: The text to be generated by the user via the user interface of user system 102 and included in message 400.

[0095] • Message image payload 406: Image data captured by the camera device component of user system 102 or retrieved from the memory component of user system 102 and included in message 400. The image data for the sent or received message 400 may be stored in image table 316.

[0096] • Message video payload 408: Video data captured by the camera device component or retrieved from the memory component of the user system 102 and included in the message 400. The video data for the sent or received message 400 may be stored in the image table 316.

[0097] • Message audio payload 410: Audio data captured by the microphone or retrieved from the memory component of the user system 102 and included in message 400.

[0098] • Message enhancement data 412: Enhancement data (e.g., filters, labels, 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 312.

[0099] • 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 made accessible to the user.

[0100] • Message geolocation parameter 416: Geolocation data (e.g., latitude and longitude coordinates) associated with the message's content payload. 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 within the message image payload 406 or a specific video within the message video payload 408).

[0101] • Message Story Identifier 418: An identifier value that identifies one or more sets of content (e.g., “story” identified in set table 318) associated with a specific content item in the message image payload 406 of message 400. For example, the identifier value can be used to associate multiple images within the message image payload 406 with multiple sets of content, respectively.

[0102] • 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. The tag value can be manually generated based on user input, or it can be automatically generated using, for example, image recognition.

[0103] • Message sender identifier 422: An identifier (e.g., message sending system identifier, email address, or device identifier) ​​indicating the user of the user system 102 on which message 400 is generated and from which message 400 is sent.

[0104] • Message receiver identifier 424: An identifier (e.g., message sending and receiving system identifier, email address, or device identifier) ​​indicating the user of the user system 102 to which message 400 is addressed.

[0105] 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, image values ​​in message image payload 406 can be pointers (or addresses) to locations within image table 316. Similarly, values ​​in message video payload 408 can point to data stored in image table 316, values ​​in message enhancement data 412 can point to data stored in enhancement table 312, values ​​in message story identifier 418 can point to data stored in set table 318, and values ​​in message sender identifier 422 and message receiver identifier 424 can point to user records stored in entity table 308.

[0106] Mesh generation system

[0107] Figure 5This is a block diagram illustrating an example mesh generation system 500 according to some examples. The mesh generation system 500 receives a 2D image depicting an object such as a human and uses a combination of multiple machine learning models to generate a 3D mesh representing that object. Specifically, the mesh generation system 500 uses a first machine learning model to simultaneously or together predict a volumetric reconstruction tensor and a pose output representing the pose of the object depicted in the image. The mesh generation system 500 accesses a canonical body reconstruction (e.g., a canonical pose) in canonical space and uses the pose output to identify pixels or points in the volumetric reconstruction tensor corresponding to specific points in the canonical pose. The mesh generation system 500 then classifies the identified pixels or points as inside or outside the canonical pose. After iteratively accessing each single point in the canonical pose in this manner, the mesh generation system 500 classifies all points in the volumetric reconstruction tensor as inside or outside the canonical pose. Points determined to be inside the canonical pose are then used to generate a 3D mesh that can be animated to provide one or more XR experiences.

[0108] In some examples, the mesh generation system 500 accesses a monocular image depicting an object. The mesh generation system 500 predicts both the volumetric reconstruction tensor of the monocular image and the pose of the object by applying a first machine learning model to the monocular image. The mesh generation system 500 identifies portions of the object's pose corresponding to points in the canonical space associated with a set of position-encoded information. The mesh generation system 500 obtains points corresponding to the identified pose portions in the volumetric reconstruction tensor. The mesh generation system 500 classifies the obtained points as either inside or outside the canonical volume by applying a second machine learning model, along with the set of position-encoded information, to the points in the obtained volumetric reconstruction tensor. In response to classifying the obtained points as inside or outside the canonical volume, the mesh generation system 500 generates a 3D mesh representing the object in the canonical space.

[0109] In some examples, the object's pose includes a skeleton representation of the object. In some examples, the mesh generation system 500 selects points in the canonical space associated with a set of position-encoded information. In some examples, the mesh generation system 500 selects a second portion of the object's pose corresponding to a second point in the canonical space associated with a second set of position-encoded information. The mesh generation system 500 obtains a second point in the volume reconstruction tensor corresponding to the second portion of the pose. The mesh generation system 500 classifies the second point as being inside or outside the canonical volume by applying a second machine learning model, along with the second set of position-encoded information, to the second point in the volume reconstruction tensor.

[0110] In some examples, the mesh generation system 500 repeatedly selects the pose portion for each part of the canonical volume and repeatedly classifies the points corresponding to the volume reconstruction tensor of the selected pose portion as either inside or outside the canonical volume. In some examples, the set of position-encoded information represents different body parts of the canonical volume, such as arms, legs, torso, head, hands, feet, etc.

[0111] In some examples, the mesh generation system 500 trains a first machine learning model and a second machine learning model by performing training operations. These training operations include obtaining a set of training images depicting objects. Each of the training images is associated with a 3D scan of the object depicted in the set of training images, and a ground-valued canonical space representation of the object depicted in the set of training images is generated by modifying the pose of the 3D scan of the object to correspond to a pose in the canonical space.

[0112] In some examples, the training operation includes selecting a first training image depicting a first object, which corresponds to a first ground-value canonical space representation in the ground-value canonical space representation of the first object. The training operation includes predicting both a single volumetric reconstruction tensor of the monocular image and a single pose of the first object by applying a first machine learning model to the first training image. The training operation includes identifying a single portion of the single pose of the first object, which corresponds to a single point in the canonical space associated with a single set of position-encoded information. The training operation includes obtaining a single point in the single volumetric reconstruction tensor corresponding to the single portion. The training operation includes classifying the single point as inside or outside a single canonical volume by applying a second machine learning model, along with the single set of position-encoded information, to the single point in the volumetric reconstruction tensor. The training operation includes generating an occupancy loss based on the deviation between the classification obtained from classifying the single point and the corresponding portion of the first ground-value canonical space representation of the object. In some examples, the mesh generation system 500 updates one or more parameters of the first and second machine learning models based on the occupancy loss.

[0113] In some examples, the training operation includes generating a single 3D mesh representing a first object in the canonical space in response to classifying a single point as inside or outside the canonical volume. The training operation includes setting the single 3D mesh based on a single pose of the first object predicted by a first machine learning model, and calculating a setting loss based on the deviation between the set single 3D mesh and a 3D scan of the first object. In some examples, the mesh generation system 500 updates one or more parameters of the first and second machine learning models based on the setting loss.

[0114] In some examples, the mesh generation system 500 crops a portion of a monocular image corresponding to the head of an object and generates a facial landmark of the object's head based on the cropped portion of the monocular image. Based on the generated facial landmark, the mesh generation system 500 predicts a head-specific canonical space reconstruction by applying a head-specific canonical network to the cropped portion. The mesh generation system 500 combines the head-specific canonical space reconstruction with the canonical volume corresponding to the object.

[0115] In some examples, the mesh generation system 500 selects facial landmarks in a head-specific canonical space representation of an object. The mesh generation system 500 identifies specific portions of the object depicted within a cropped monocular image that correspond to the selected facial landmarks. The mesh generation system 500 obtains a single point corresponding to the volumetric reconstruction tensor of the specific portion and classifies the obtained single point as either inside or outside the head-specific canonical space representation of the object by applying a third machine learning model to the obtained single point. In response to classifying the obtained single point as inside or outside the head-specific canonical space representation, the mesh generation system 500 generates a head-specific canonical space reconstruction.

[0116] In some examples, the mesh generation system 500 predicts both the additional volumetric reconstruction tensor of the cropped monocular image portion and the head pose of the object by applying a fourth machine learning model to the portion. Individual points can be obtained from the additional volumetric reconstruction tensor. In some examples, the mesh generation system 500 updates the pose of the head region in the canonical volume based on facial landmarks. The canonical volume with the updated pose of the head region is used to generate a 3D mesh.

[0117] In some examples, the canonical volume represents the full-body canonical space reconstruction of the object. The mesh generation system 500 identifies the region where the full-body canonical space reconstruction of the object overlaps with the head-specific canonical space reconstruction. The mesh generation system 500 smoothly deforms this region based on the difference between the full-body and head-specific canonical space reconstructions (e.g., through the integral value of the region). In some examples, the mesh generation system 500 generates a 3D mesh based on the full-body canonical space reconstruction, the head-specific canonical space reconstruction, and / or the smoothly deformed region.

[0118] In some examples, the mesh generation system 500 generates XR objects based on 3D meshes. In some examples, the mesh generation system 500 animates XR objects in an XR view of a real-world environment captured in real time by a camera device of the user system 102.

[0119] For example, the mesh generation system 500 includes an image access component 510. The image access component 510 can access real-time or previously stored images or videos from the user system 102. The image access component 510 can select frames or sets of frames from the accessed images or videos to generate a 3D mesh depicting one or more real-world or XR objects within the selected frames or set of frames. During training of the mesh generation system 500, the image access component 510 accesses a database of training data to retrieve images depicting real-world objects and their corresponding ground truth 3D scans to train one or more machine learning models for the mesh generation system 500.

[0120] During operation (e.g., after training), the image access component 510 simultaneously or sequentially provides images of the depicted object to the body canonical reconstruction network 520 and / or to the head-specific canonical reconstruction network 530. The body canonical reconstruction network 520 includes a plurality of machine learning models (e.g., a first machine learning model and a second machine learning model) trained to process the received images and generate canonical representations of the objects depicted in the received images. The following is in conjunction with... Figure 6 The operations performed by the body canonical reconstruction network 520 are described in more detail. The head-specific canonical reconstruction network 530 includes a second plurality of machine learning models (e.g., a third and a fourth machine learning model) trained to process the received image, for example, to crop portions of the image corresponding to the head based on the outputs of the first and / or second machine learning models, and to generate a head-specific canonical representation of only the head portion of the object depicted in the received image. In this way, the body canonical reconstruction network 520 provides or generates a general and universal body representation of the object depicted in the image in canonical space, and the head-specific canonical reconstruction network 530 generates a head-specific representation of the facial features and head of the object depicted in the image, which includes a very detailed representation of the object's head and face.

[0121] The outputs of the head-specific canonical reconstruction network 530 and the body canonical reconstruction network 520 are provided to the mesh generation component 540. The mesh generation component 540 combines the body portions of the object's body representation (e.g., excluding the head) with the head portions of the object's head-specific representation (e.g., excluding the neck, which may be included in the output of the head-specific canonical reconstruction network 530). The mesh generation component 540 can identify regions of the body representation that overlap with regions of the head-specific representation. For example, the mesh generation component 540 can determine that the neck in the head-specific representation is also included in the body representation. This overlapping region can be interpolated or smoothed to smooth any differences between the overlapping regions or to interpolate them. The mesh generation component 540 then includes this smoothed region in the combined reconstruction of the object formed by the body representation and the head-specific representation to form a final representation of the object depicted in the image. The mesh generation component 540 uses the final representation of the object to generate a 3D mesh and applies this 3D mesh to the character or avatar depicted in the image to provide an XR experience. Mesh generation component 540 animates characters or avatars based on the movement of people depicted in an image.

[0122] Figure 6 This is a graphical representation of a full-body canonical space reconstruction system 600 based on some examples. Specifically, the full-body canonical space reconstruction system 600 may implement some or all of the components of a body canonical reconstruction network 520. The full-body canonical space reconstruction system 600 includes an image 601 that can be received from an image access component 510. Image 601 may depict objects, such as people and backgrounds. The full-body canonical space reconstruction system 600 includes a first machine learning model 610. The first machine learning model 610 processes image 601 and simultaneously outputs both a volumetric reconstruction tensor 612 and a pose output 614. The volumetric reconstruction tensor 612 is a tensor of estimated pixel values ​​for each point of image 601. In some cases, the volumetric reconstruction tensor 612 provides estimated pixel values ​​or estimated 3D points for various portions of image 601 in 3D. The volumetric reconstruction tensor 612 may be a mathematical representation of a 3D scene or object using a four-dimensional (4D) tensor, which is a multidimensional array of numbers. This tensor can capture spatial and angular information about a scene or object, such as its shape, color, and lighting.

[0123] Pose output 614 includes a skeleton with various bones arranged to represent the current pose of the object depicted in image 601. The bones of the skeleton included in pose output 614 can be arranged to mimic the appearance of the skeletal features of the object (e.g., a person) depicted in image 601. In some examples, the full-body canonical space reconstruction system 600 accesses a canonical pose 640 representing the pose of the object in canonical space. In some cases, canonical pose 640 may correspond to the object's T-pose, where the arms are extended as far away from the object's body extensions as possible and the legs are also extended. Canonical pose 640 may differ from pose output 614. In some cases, the canonical pose may be any other type of pose and is not limited to the T-pose.

[0124] The full-body canonical space reconstruction system 600 selects a specific portion of the canonical posture 640, such as the torso or hand portion of the canonical posture 640. The full-body canonical space reconstruction system 600 identifies a set of positional coding information corresponding to the selected specific portion. The set of positional coding information can specify a set of points and their positions for the specific portion. That is, a first set of positional coding information can identify or specify the positions of points corresponding to a first body part (e.g., the torso) in the canonical posture 640. A second set of positional coding information can identify or specify the positions of points corresponding to a second body part (e.g., the legs) in the canonical posture 640.

[0125] The full-body canonical space reconstruction system 600 can select specific points from a specific portion of the canonical posture 640. The full-body canonical space reconstruction system 600 can then access posture output 614 to identify portions of posture output 614 corresponding to the selected specific portion of the canonical posture 640. For example, if the torso is selected from the canonical posture 640, the full-body canonical space reconstruction system 600 can identify the torso region of posture output 614. The full-body canonical space reconstruction system 600 can select or identify points within posture output 614 corresponding to points in the selected specific portion of the canonical posture 640.

[0126] The full-body canonical spatial reconstruction system 600 uses points from the selected pose output 614 to retrieve points or sets of points from the volumetric reconstruction tensor 612 that match the points of the selected pose output 614. In some cases, the full-body canonical spatial reconstruction system 600 repeats the above processing for each point and each portion of the canonical pose 640 to form a signal 620 including an image alignment tensor and position-coded values ​​corresponding to a specific portion of each selected canonical pose 640. For example, the full-body canonical spatial reconstruction system 600 can select a set of points from a second portion of the canonical pose 640 associated with a second set of position-coded information. The full-body canonical spatial reconstruction system 600 can identify which portion of the pose output 614 corresponds to the second portion of the canonical pose 640 and use that portion to retrieve pixel values ​​from the volumetric reconstruction tensor 612. The retrieved pixel values ​​are then stored in the signal 620 in association with the second set of position-coded information. The signal 620 is generated after processing all portions of the canonical pose 640 in this way to identify, retrieve each pixel in the volumetric reconstruction tensor 612 and associate it with the corresponding position-coded information.

[0127] The full-body canonical spatial reconstruction system 600 can then provide a signal 620 to a second machine learning model 630, which includes image alignment tensors and positional encoding values ​​corresponding to one or more portions (or all portions) of the canonical pose 640. The second machine learning model 630 classifies each point in the signal 620 as either inside or outside the canonical pose 640. After all points in the volumetric reconstruction tensor 612 have been classified, the output of the full-body canonical spatial reconstruction system 600 is the pixel values ​​classified as inside the canonical pose 640 to form a body representation of the object depicted in image 601. That is, this body representation corresponds to the output of the body canonical reconstruction network 520.

[0128] In some examples, the whole-body canonical space reconstruction system 600 is trained using training data received from the image access component 510. Various types of training losses, including template loss and setting loss, can be used to train the whole-body canonical space reconstruction system 600. Figure 7 It is a graphical representation of the training loss of a whole-body canonical space reconstruction system 600 based on some examples 700.

[0129] like Figure 7As shown, to train the full-body canonical space reconstruction system 600, the full-body canonical space reconstruction system 600 retrieves a first training image depicting a person and a corresponding 3D scan of that person. The full-body canonical space reconstruction system 600 uses the 3D scan to generate a ground truth canonical pose representing the person. The full-body canonical space reconstruction system 600 then processes the first training image using a first machine learning model 610 to generate a training volumetric reconstruction tensor 612 and a corresponding pose output 614 representing the current pose of the person depicted in the first training image. The full-body canonical space reconstruction system 600 selects points along the canonical pose (e.g., representing the ground truth canonical pose of the person) and uses those points to retrieve a set of corresponding points from the volumetric reconstruction tensor 612 based on the pose output 614.

[0130] That is, the full-body canonical space reconstruction system 600 forms a signal representing all points from the volume reconstruction tensor 612, all points being associated with different parts of the canonical pose. To this end, the full-body canonical space reconstruction system 600 can select the arm portion of the canonical pose and use the pose output 614 to determine which part or set of pixels in the volume reconstruction tensor 612 corresponds to the arm portion. Those pixels are then inserted into the signal and associated with the arm portion. This process is repeated for the leg portion, torso portion, and all other parts of the canonical pose until all parts are associated with (e.g., labeled) location encoding information of the corresponding pixel values ​​in the volume reconstruction tensor 612. The full-body canonical space reconstruction system 600 then applies a second machine learning model 630 to the signal including the pixels and associated location encoding information to classify pixel values ​​as inside or outside the canonical space. The full-body canonical space reconstruction system 600 retrieves pixels classified as inside the canonical space and compares those values ​​with ground truth pixel values ​​from the ground truth canonical space representation of the person. The full-body canonical space reconstruction system 600 calculates the deviation between pixels classified as falling within the canonical space and ground-value pixel values ​​from the ground-value canonical space representation of the human body. This deviation is used to generate an occupancy loss (e.g., template space reconstruction loss 710), which is used to update one or more parameters of the first machine learning model 610 and / or the second machine learning model 630. In this way, the first machine learning model 610, which generates both the pose output 614 and the volume reconstruction tensor 612, is updated using the same training loss, which improves the overall accuracy and efficiency of generating these signals.

[0131] In some examples, the full-body canonical space reconstruction system 600 uses pixels classified as being within canonical space to generate a 3D training mesh. The 3D training mesh is set based on the pose output 614 of a first training image. The 3D training mesh is then compared to a 3D scan of the person depicted in the first training image to generate a set-space reconstruction loss 720. The bias resulting from this comparison is used to generate the set loss, which is used to update one or more parameters of a first machine learning model 610 and / or a second machine learning model 630. In this way, the first machine learning model 610, which generates both the pose output 614 and the volume reconstruction tensor 612, is updated using this additional training loss, which improves the overall accuracy and efficiency of generating these signals.

[0132] Figure 8 This is a detailed graphical representation 800 of the head pose of a full-body canonical space reconstruction system based on some examples. Specifically, the head-specific canonical reconstruction network 530 can process images received from the image access unit 510 to crop regions 810 of the image corresponding to the head portion of the object depicted in the image. The head-specific canonical reconstruction network 530 can use the pose output 614 received from the signal 620 to select the portion of the image corresponding to the head and / or can use any other image recognition technique to retrieve and identify the portion of the image corresponding to the head.

[0133] A head-specific canonical reconstruction network 530 can process the cropped region 810 of the image to identify 2D facial landmarks of the head depicted in the cropped region 810. These 2D facial landmarks can be fed to a body canonical reconstruction network 520 to update the head pose generated by the body canonical reconstruction network 520. That is, the output of the body canonical reconstruction network 520 can be used to generate a 3D mesh, such as mesh 820. Mesh 820 can then be updated, for example, by thinning or modifying the pose of the head portion to generate a new mesh 830, in which the head pose is modified based on the 2D facial landmarks detected in the cropped region 810.

[0134] In some examples, the head-specific canonical reconstruction network 530 can process the cropped region 810 in parallel with or sequentially after the body canonical reconstruction network 520 for generating canonical body representations of objects depicted in images received from the image access component 510. The head-specific canonical reconstruction network 530 can perform similar processing as the body canonical reconstruction network 520 to generate a head-specific canonical representation of the head depicted in the cropped region 810. For example, the head-specific canonical reconstruction network 530 can implement a third machine learning model (which can perform similar functions to the first machine learning model 610, but specific to the human head portion). The third machine learning model processes the cropped region 810 and simultaneously outputs both a head-specific volumetric reconstruction tensor and a head-specific pose output.

[0135] In some examples, the head-specific canonical reconstruction network 530 accesses a head-specific canonical pose representing the pose of an object's head in the canonical space. The head-specific canonical reconstruction network 530 selects facial features specific to the head-specific canonical pose, such as the eyes or mouth portion. The head-specific canonical reconstruction network 530 identifies a set of head-specific positional encoding information corresponding to the selected facial features. The set of positional encoding information can specify a set of points and their locations for a specific facial feature. The head-specific canonical reconstruction network 530 can then access a head-specific pose output 614 to identify portions of the head-specific pose output 614 corresponding to the selected head-specific canonical pose. The head-specific canonical reconstruction network 530 uses the points from the selected head-specific pose output to retrieve points or sets of points from the head-specific volume reconstruction tensor that match the points from the selected head-specific pose output. In some cases, the head-specific canonical reconstruction network 530 repeats the above processing for each point and each facial feature of the head-specific canonical pose to form a signal that includes image alignments of specific facial features corresponding to each selected head-specific canonical pose, along with head-specific tensor and positional encoding values. After processing all facial features of the head-specific canonical pose in this way to identify, retrieve, and associate each pixel in the head-specific volumetric reconstruction tensor with its corresponding positional encoding information, a head-specific signal is generated.

[0136] The head-specific canonical reconstruction network 530 can then provide a head-specific signal to the fourth machine learning model, which includes head-specific tensor and positional encoding values ​​corresponding to one or more facial features (or all facial features) in an image alignment specific to a head-specific canonical pose. The fourth machine learning model classifies each point in the head-specific signal as either inside or outside the head-specific canonical pose. After all points in the head-specific volumetric reconstruction tensor have been classified, the head-specific canonical reconstruction network 530 outputs pixel values ​​classified as inside the head-specific canonical pose to form a head-specific representation of the object depicted in the image.

[0137] In some examples, a head-specific canonical space reconstruction system is trained using training data that is the same as or different from the training data used to train the whole-body canonical space reconstruction system 600. To train the head-specific canonical space reconstruction system, it retrieves a first training image depicting a human head and a corresponding 3D scan of the person. The head-specific canonical space reconstruction system uses the 3D scan to generate a ground truth canonical head-specific pose representing the human head. The head-specific canonical space reconstruction system then processes the first training image using a third machine learning model to generate a training head-specific volumetric reconstruction tensor and a corresponding head-specific pose output representing the current head-specific pose of the person depicted in the first training image. The head-specific canonical space reconstruction system selects points along the head-specific canonical pose (e.g., representing the ground truth canonical pose of the human head) and uses those points to retrieve a set of corresponding points from the head-specific volumetric reconstruction tensor based on the head-specific pose output 614.

[0138] The head-specific canonical space reconstruction system then applies a fourth machine learning model to the signal, which includes head-specific pixels and associated location-encoded information, to classify pixel values ​​as either inside or outside the head-specific canonical space. The head-specific canonical space reconstruction system retrieves pixels classified as within the head-specific canonical space and compares those values ​​to ground truth pixel values ​​from the human's ground truth head-specific canonical space representation. The head-specific canonical space reconstruction system calculates the deviation between pixels classified as within the head-specific canonical space and the ground truth pixel values ​​from the human's ground truth head-specific canonical space representation. This deviation is used to generate a head-specific occupancy loss, which is used to update one or more parameters of the third and / or fourth machine learning models.

[0139] Figure 9 This is a graphical representation 900 of the mesh generation components of a whole-body canonical space reconstruction system based on some examples. Specifically, such as Figure 9As shown, input image 601 can be processed by body canonical reconstruction network 520 and head-specific canonical reconstruction network 530 to generate a head-specific canonical representation 910 and a full-body canonical representation 920 of the object depicted in image 601. Mesh generation system 500 can use the head portion of the head-specific canonical representation 910 and the body portion of the full-body canonical representation 920 to generate a final reconstruction 930 of the object depicted in image 601. Specifically, mesh generation system 500 can select a threshold region of the head-specific canonical representation 910 (e.g., the top of a person's neck) used to cut off the portion of the head-specific canonical representation 910 used to form the final reconstruction 930. Similarly, mesh generation system 500 can select another threshold region of the full-body canonical representation 920 (e.g., the bottom of a person's neck) used to cut off the portion of the full-body canonical representation 920 used to form the final reconstruction 930. The mesh generation system 500 can determine the overlap between region 912 of the head-specific canonical representation 910 and region 922 of the whole-body canonical representation 920. That is, both the head-specific canonical representation 910 and the whole-body canonical representation 920 can include the neck region. The mesh generation system 500 can interpolate the differences between the overlapping portions to smooth the pixel values ​​and use those smoothed pixel values ​​to generate a final reconstruction 930. The final reconstruction 930 can then be used to form a 3D mesh of the object depicted in image 601.

[0140] Figure 10 This is a flowchart of a process 1000 performed by a grid generation system 500, based on some examples. Although the flowchart describes the operations as sequential processes, many operations within an operation can be performed in parallel or simultaneously. Furthermore, the order of operations can be rearranged. A process terminates when its operations are complete. A process can correspond to a method, procedure, etc. The steps of a method can be performed in whole or in part, in combination with some or all steps of other methods, and can be performed by any number of different systems or any part thereof (e.g., a processor included in any system within the system).

[0141] At operation 1001, as discussed above, the mesh generation system 500 (e.g., user system 102 or server) accesses a monocular image of the depicted object.

[0142] At operation 1002, as discussed above, the mesh generation system 500 predicts both the volumetric reconstruction tensor of the monocular image and the pose of the object by applying a first machine learning model to the monocular image.

[0143] At operation 1003, as discussed above, the mesh generation system 500 identifies a portion of the pose of an object corresponding to a point in the canonical space associated with a set of location-encoded information.

[0144] At operation 1004, as discussed above, the mesh generation system 500 obtains the point of the volume reconstruction tensor corresponding to the portion of the identified pose.

[0145] At operation 1005, as discussed above, the mesh generation system 500 classifies the obtained points as inside or outside the canonical volume by applying a second machine learning model together with a set of location encoding information to the points of the obtained volume reconstruction tensor.

[0146] At operation 1006, as discussed above, the mesh generation system 500 generates a 3D mesh representing the object in the canonical space in response to classifying the obtained points as inside or outside the canonical volume.

[0147] Example

[0148] Example 1. A method comprising: accessing a monocular image depicting an object; predicting both a volumetric reconstruction tensor of the monocular image and the pose of the object by applying a first machine learning model to the monocular image; identifying a portion of the object's pose corresponding to points in a canonical space associated with a set of positional encoding information; obtaining points in the volumetric reconstruction tensor corresponding to the identified portion of the pose; classifying the obtained points as inside or outside the canonical volume by applying a second machine learning model together with the set of positional encoding information to the points in the obtained volumetric reconstruction tensor; and generating a 3D mesh representing the object in the canonical space in response to classifying the obtained points as inside or outside the canonical volume.

[0149] Example 2. According to the method described in Example 1, the pose of the object includes a skeleton representation of the object.

[0150] Example 3. The method according to any one of Examples 1 to 2 further includes: selecting points in the canonical space associated with a set of location coding information.

[0151] Example 4. The method according to any one of Examples 1 to 3 further includes: selecting a second part of the pose of an object corresponding to a second point associated with a second set of position encoding information in the canonical space; obtaining a second point of the volume reconstruction tensor corresponding to the second part of the pose; and classifying the second point as inside or outside the canonical volume by applying a second machine learning model together with the second set of position encoding information to the second point of the volume reconstruction tensor.

[0152] Example 5. The method according to Example 4 further includes, for each part of the canonical volume, repeatedly selecting the part of the pose, and repeatedly classifying the points of the volume reconstruction tensor corresponding to the part of the selected pose as inside or outside the canonical volume.

[0153] Example 6. The method according to any one of Examples 1 to 5, wherein the set of location-encoded information represents different body parts of the canonical volume.

[0154] Example 7. The method according to any one of Examples 1 to 6 further includes training a first machine learning model and a second machine learning model by performing a training operation, the training operation comprising: obtaining a set of training images depicting an object, each of the training images being associated with a 3D scan of the object depicted in the set of training images; and generating a ground truth canonical space representation of the object depicted in the set of training images by modifying the pose of the 3D scan of the object to correspond to a pose in canonical space.

[0155] Example 8. According to the method of Example 7, wherein the training operation includes: selecting a first training image depicting a first object, the first training image corresponding to a first ground-value canonical space representation in the ground-value canonical space representation of the first object; predicting both a single volumetric reconstruction tensor of a monocular image and a single pose of the first object by applying a first machine learning model to the first training image; identifying a single portion of the single pose of the first object corresponding to a single point in the canonical space associated with a single set of positional encoding information; obtaining a single point in the single volumetric reconstruction tensor corresponding to the single portion; classifying the single point as inside or outside a single canonical volume by applying a second machine learning model together with the single set of positional encoding information to the single point in the volumetric reconstruction tensor; and generating an occupancy loss based on the deviation between the classification obtained by classifying the single point and the corresponding portion of the first ground-value canonical space representation of the object.

[0156] Example 9. The method described in Example 8 further includes updating one or more parameters of the first machine learning model and the second machine learning model based on occupancy loss.

[0157] Example 10. The method according to any one of Examples 8 to 9 further includes: generating a single 3D mesh representing a first object in a canonical space in response to classifying the single point as inside or outside the canonical volume; setting the single 3D mesh based on a single pose of the first object predicted by a first machine learning model; and calculating a setting loss based on the deviation between the set single 3D mesh and a 3D scan of the first object.

[0158] Example 11. The method according to any one of Examples 1 to 10 further includes: cropping a portion of a monocular image corresponding to the head of an object; generating a facial landmark of the head of the object based on the cropped portion of the monocular image; predicting a head-specific canonical space reconstruction based on the generated facial landmark by applying a head-specific canonical network to the cropped portion; and combining the head-specific canonical space reconstruction with a canonical volume corresponding to the object.

[0159] Example 12. The method according to Example 11 further includes: selecting a facial landmark in a head-specific canonical space representation of an object; identifying a specific portion of the object depicted in a cropped monocular image corresponding to the selected facial landmark; obtaining a single point corresponding to a volumetric reconstruction tensor of the specific portion; classifying the obtained single point as inside or outside the head-specific canonical space representation of the object by applying a third machine learning model to the obtained single point; and generating a head-specific canonical space reconstruction in response to classifying the obtained single point as inside or outside the head-specific canonical space representation.

[0160] Example 13. The method according to Example 12 further includes: predicting both an additional volumetric reconstruction tensor of the cropped monocular image portion and the head pose of the object by applying a fourth machine learning model to the cropped monocular image portion, wherein a single point is obtained from the additional volumetric reconstruction tensor.

[0161] Example 14. The method according to any one of Examples 11 to 13 further includes: updating the pose of the head region in the canonical volume based on facial landmarks, wherein the canonical volume with the updated pose of the head region is used to generate a 3D mesh.

[0162] Example 15. The method according to any one of Examples 11 to 14, wherein the canonical volume represents a full-body canonical space reconstruction of an object, further includes: identifying a region where the full-body canonical space reconstruction of the object overlaps with a head-specific canonical space reconstruction; and smoothly deforming the region based on the difference between the full-body canonical space reconstruction of the object and the head-specific canonical space reconstruction.

[0163] Example 16. The method described in Example 15 further includes generating a 3D mesh based on whole-body canonical space reconstruction, head-specific canonical space reconstruction, and smoothly deformed regions.

[0164] Example 17. The method according to any one of Examples 1 to 16 further includes generating XR objects based on 3D meshes.

[0165] Example 18. The method described in Example 17 further includes animates the XR object in a view of a real-world environment captured in real time by a camera device of the user system.

[0166] Example 19. A system comprising: at least one processor; and at least one memory unit having instructions stored thereon, the instructions causing the at least one processor to perform operations when executed by the at least one processor, the operations including: accessing a monocular image depicting an object; predicting both a volumetric reconstruction tensor of the monocular image and the pose of the object by applying a first machine learning model to the monocular image; identifying a portion of the pose of the object corresponding to points in a canonical space associated with a set of positional encoding information; obtaining points of the volumetric reconstruction tensor corresponding to the identified portion of the pose; classifying the obtained points as inside or outside the canonical volume by applying a second machine learning model together with the set of positional encoding information to the points of the obtained volumetric reconstruction tensor; and generating a 3D mesh representing the object in the canonical space in response to classifying the obtained points as inside or outside the canonical volume.

[0167] Example 20. A non-transitory computer-readable storage medium having instructions stored thereon, the instructions causing at least one processor to perform operations when executed by at least one processor, the operations including: accessing a monocular image depicting an object; predicting both a volumetric reconstruction tensor of the monocular image and the pose of the object by applying a first machine learning model to the monocular image; identifying a portion of the pose of the object corresponding to points in a canonical space associated with a set of position-encoded information; obtaining points of the volumetric reconstruction tensor corresponding to the identified portion of the pose; classifying the obtained points as inside or outside the canonical volume by applying a second machine learning model together with the set of position-encoded information to the points of the obtained volumetric reconstruction tensor; and generating a 3D mesh representing the object in the canonical space in response to classifying the obtained points as inside or outside the canonical volume.

[0168] Machine architecture

[0169] Figure 11This is a schematic representation of machine 1100, within which instructions 1102 (e.g., software, programs, applications, applets, or other executable code) can be executed to cause machine 1100 to perform any or more of the methods discussed herein. For example, instructions 1102 can cause machine 1100 to perform any or more of the methods described herein. Instructions 1102 transform the general, unprogrammed machine 1100 into a specific machine 1100 programmed to perform the described and illustrated functions in the described manner. Machine 1100 can operate as a standalone device or can be coupled (e.g., networked) to other machines. In a networked deployment, machine 1100 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 1100 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 1102 specifying actions to be taken by machine 1100. Furthermore, although only a single machine 1100 is shown, the term "machine" should also be considered as a collection of machines that individually or jointly execute instructions 1102 to perform any or more of the methods discussed herein. For example, machine 1100 may include user system 102 or any of a plurality of server devices forming part of interactive server system 110. In some examples, machine 1100 may also include both client and server systems, wherein certain operations of a particular method or algorithm are performed on the server side and certain operations of said particular method or algorithm are performed on the client side.

[0170] Machine 1100 may include a processor 1104, a memory 1106, and an input / output (I / O) unit 1108 that can be configured to communicate with each other via a bus 1110. In the example, processor 1104 (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, processors 1112 and 1114 that execute instruction 1102. 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 11 Multiple processors 1104 are shown, but machine 1100 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.

[0171] Memory 1106 includes main memory 1116, static memory 1118, and storage cells 1120, all of which are accessible by processor 1104 via bus 1110. Main memory 1106, static memory 1118, and storage cells 1120 store instructions 1102 that implement any one or more of the methods or functions described herein. Instructions 1102 may also reside wholly or partially in main memory 1116, static memory 1118, machine-readable medium 1122 within storage cell 1120, at least one processor of processor 1104 (e.g., in the processor's cache memory), or any suitable combination thereof during execution by machine 1100.

[0172] I / O component 1108 may include various components for receiving input, providing output, generating output, sending information, exchanging information, capturing measurement results, etc. The specific I / O component 1108 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 is unlikely to include such a touch input device. It should be recognized that I / O component 1108 may include... Figure 11Many other components are not shown. In various examples, I / O component 1108 may include user output component 1124 and user input component 1126. User output component 1124 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 1126 may include alphanumeric input components (e.g., keyboards, touchscreens configured to receive alphanumeric input, photoelectric 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 the position and force of a touch or touch gesture), audio input components (e.g., microphones), etc. Any biometric data collected by the biometric component is captured and stored with user approval and deleted upon user request.

[0173] Furthermore, such biometric data can be used for very limited purposes, such as identity verification. To ensure limited and authorized use of biometric information and other personally identifiable information (PII), access to the data is restricted to authorized personnel (if permitted). Any use of biometric data can be strictly limited to identity verification purposes, and the data will not be shared or sold to any third party without the user's explicit consent. Additionally, appropriate technical and organizational measures are implemented to ensure the security and confidentiality of this sensitive information.

[0174] In other examples, I / O component 1108 may include biometric component 1128, motion component 1130, environmental component 1132, or position component 1134, as well as a wide range of other components. For example, biometric component 1128 includes components for detecting expressions (e.g., hand gestures, facial expressions, voice expressions, body posture, or eye tracking), measuring biosignals (e.g., blood pressure, heart rate, body temperature, sweating, or brainwaves), and identifying a person (e.g., voice recognition, retinal recognition, facial recognition, fingerprint recognition, or EEG-based recognition). Biometric components may include brain-computer interface (BMI) systems that allow communication between the brain and external devices or machines. This can be achieved by recording brain activity data, converting that data into a format that can be understood by a computer, and then using the resulting signals to control a device or machine.

[0175] Examples of BMI technology types include:

[0176] • Brain-based brain-computer interfaces (BMIs) use electrodes placed on the scalp to record electrical activity in the brain.

[0177] • Invasive BMI, which uses electrodes surgically implanted into the brain.

[0178] • Optogenetics BMI, which uses light to control the activity of specific nerve cells in the brain.

[0179] The moving part 1130 includes an acceleration sensor part (e.g., an accelerometer), a gravity sensor part, and a rotation sensor part (e.g., a gyroscope).

[0180] The environmental component 1132 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 for safety purposes or for measuring pollutants in the atmosphere), or other components that can provide indications, measurements, or signals corresponding to the surrounding physical environment.

[0181] Regarding the camera device, user system 102 may have a camera device system including, for example, a front-facing camera on the front surface of user system 102 and a rear-facing camera on the rear surface of user system 102. The front-facing camera may be used, for example, to capture still images and videos (e.g., "selfies") of the user of user system 102, which can then be enhanced with the aforementioned enhancement data (e.g., filters). For example, the rear-facing camera may be used 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, user system 102 may also include a 360° camera for capturing 360° photos and videos.

[0182] Furthermore, the camera system of user system 102 may include dual rear cameras (e.g., a main camera and a depth-sensing camera), or even triple, quadruple, or quintuple rear camera configurations on the front and rear sides of user system 102. For example, these multi-camera systems may include wide-angle cameras, ultra-wide-angle cameras, telephoto cameras, macro cameras, and depth sensors.

[0183] The position component 1134 includes a position sensor component (e.g., a GPS receiver component), an altitude sensor component (e.g., an altimeter or barometer that detects air pressure and from which altitude can be obtained), an orientation sensor component (e.g., a magnetometer), etc.

[0184] Various technologies can be used to implement communication. I / O component 1108 also includes a communication component 1136 operable to couple machine 1100 to network 1138 or device 1140 via a suitable coupling or connection. For example, communication component 1136 may include a network interface component or other suitable device that interfaces with network 1138. In other examples, communication component 1136 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 for providing communication via other forms. Device 1140 may be another machine or any peripheral device from a variety of peripheral devices (e.g., a peripheral device coupled via Universal Serial Bus (USB)).

[0185] Furthermore, communication component 1136 can detect identifiers, or includes components operable to detect identifiers. For example, communication component 1136 may include a radio frequency identification (RFID) tag reader component, an NFC smart tag detection component, or an optical reader component (e.g., for detecting one-dimensional barcodes such as Universal Product Code (UPC) barcodes, QR codes such as Quick Response (QR) codes, Aztec codes, data matrices, Dataglyphs, etc.). TM The system can utilize optical sensors for multidimensional barcodes and other optical codes, such as MaxiCode, PDF417, UltraCode, and UCC RSS-2D barcodes, or acoustic detection components (e.g., microphones for identifying audio signals of the tags). Additionally, various information can be obtained via communication component 1136, such as location via Internet Protocol (IP) geolocation, location via Wi-Fi® signal triangulation, and location via detecting NFC beacon signals that indicate a specific location.

[0186] Various memories (e.g., main memory 1116, static memory 1118, and the memory of processor 1104) and storage unit 1120 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 1102) cause various operations to implement the disclosed examples when executed by processor 1104.

[0187] Instructions 1102 can be sent or received over network 1138 via a transmission medium using a network interface device (e.g., a network interface component included in communication component 1136) and using any of several known transmission protocols (e.g., HTTP). Similarly, instructions 1102 can be sent or received via a transmission medium coupled to device 1140 (e.g., peer-to-peer coupling).

[0188] Software Architecture

[0189] Figure 12 This is a block diagram 1200 illustrating a software architecture 1202 that can be installed on any or more of the devices described herein. The software architecture 1202 is supported by hardware, such as a machine 1204 including a processor 1206, memory 1208, and I / O components 1210. In this example, the software architecture 1202 can be conceptualized as a stack of layers, where each layer provides specific functionality. The software architecture 1202 includes layers such as an operating system 1212, libraries 1214, frameworks 1216, and applications 1218. In operation, application 1218 activates API calls 1220 via the software stack and receives messages 1222 in response to API calls 1220.

[0190] Operating system 1212 manages hardware resources and provides public services. Operating system 1212 includes, for example, a kernel 1224, services 1226, and drivers 1228. Kernel 1224 serves as an abstraction layer between hardware and other software layers. For example, kernel 1224 provides functions such as memory management, processor management (e.g., scheduling), component management, networking, and security settings. Services 1226 can provide other public services to other software layers. Drivers 1228 are responsible for controlling or interfacing with the underlying hardware. For example, drivers 1228 may include display drivers, camera drivers, BLUETOOTH® or BLUETOOTH® low-power drivers, flash memory drivers, serial communication drivers (e.g., USB drivers), Wi-Fi® drivers, audio drivers, power management drivers, etc.

[0191] Library 1214 provides common low-level infrastructure used by application 1218. Library 1214 may include system libraries 1230 (e.g., the C standard library), which provide functions such as memory allocation, string manipulation, and mathematical functions. Furthermore, library 1214 may include API libraries 1232, 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 codecs, Joint Picture Experts Group (JPEG or JPG), or Portable Web Graphics (PNG)), graphics libraries (e.g., OpenGL frameworks for rendering graphic content in 2D and 3D on a display), database libraries (e.g., SQLite, which provides various relational database functions), web libraries (e.g., WebKit, which provides web browsing capabilities), etc. Library 1214 may also include a wide variety of other libraries 1234 to provide many other APIs to application 1218.

[0192] Framework 1216 provides common high-level infrastructure for use by application 1218. For example, framework 1216 provides various GUI functions, high-level resource management, and high-level location services. Framework 1216 can provide a wide range of other APIs that can be used by application 1218, some of which may be specific to a particular operating system or platform.

[0193] In the example, application 1218 may include home application 1236, contact application 1238, browser application 1240, book reader application 1242, location application 1244, media application 1246, messaging application 1248, game application 1250, and a wide variety of other applications such as third-party application 1252. Application 1218 is a program that performs the functions defined in the program. One or more applications 1218 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, third-party application 1252 (e.g., an application developed by an entity other than a platform-specific vendor using the ANDROID™ or iOS™ SDK) may be mobile software that runs on mobile operating systems such as iOS™, ANDROID®, WINDOWS® Phone, or other mobile operating systems. In this example, a third-party application 1252 can activate API call 1220 provided by the operating system 1212 to facilitate the functionality described herein.

[0194] Systems with head-mounted wearable devices

[0195] Figure 13 A system 1300 including a head-mounted wearable device 116 with a selector input device is shown according to some examples. Figure 13 This is a high-level functional block diagram of an example head-mounted wearable device 116 that is communicatively coupled to mobile device 114 and various server systems 1304 (e.g., interactive server system 110) via various networks 1316.

[0196] The head-mounted wearable device 116 includes one or more camera devices, each of which may be, for example, a visible light camera 1306, an infrared emitter 1308, and an infrared camera 1310.

[0197] Mobile device 114 connects to head-mounted wearable device 116 using both low-power wireless connection 1312 and high-speed wireless connection 1314. Mobile device 114 is also connected to server system 1304 and network 1316.

[0198] The head-mounted wearable device 116 also includes two image displays 1318 of an optical assembly. The two image displays 1318 of the optical assembly include one image display associated with the left lateral side of the head-mounted wearable device 116 and one image display associated with the right lateral side of the head-mounted wearable device 116. The head-mounted wearable device 116 also includes an image display driver 1320, an image processor 1322, a low-power circuitry system 1324, and a high-speed circuitry system 1326. The image displays 1318 of the optical assembly are used to present images and videos, including images that may include a GUI, to the user of the head-mounted wearable device 116.

[0199] The image display driver 1320 commands and controls the image display 1318 of the optical components. The image display driver 1320 can directly deliver image data to the image display 1318 of the optical components for presentation, or it can convert the image data into a signal or data format suitable for delivery to the image display device. For example, the image data can be video data formatted according to compression formats such as H.264 (MPEG-4 Part 10), HEVC, Theora, Dirac, RealVideo RV40, VP8, VP9, ​​etc., and still image data can be formatted according to compression formats such as PNG, JPEG, Tagged Image File Format (TIFF), or Exchangeable Image File Format (EXIF).

[0200] The head-mounted wearable device 116 includes a frame and a stem (or temple) extending laterally from the frame. The head-mounted wearable device 116 also includes a user input device 1328 (e.g., a touch sensor or a press button), which includes an input surface on the head-mounted wearable device 116. The user input device 1328 (e.g., a touch sensor or a press button) is used to receive input selections from the user to manipulate a GUI of the presented image.

[0201] Figure 13 The components shown for the head-mounted wearable device 116 are located on one or more circuit boards (e.g., PCBs or flexible PCBs) in the eyeglass frame or temples. Alternatively or additionally, the depicted components may be located in modules, frames, hinges, or nose bridges of the head-mounted wearable device 116. The left and right visible light imaging devices 1306 may include digital imaging device elements, such as complementary metal-oxide-semiconductor (CMOS) image sensors, charge-coupled devices, camera lenses, or any other corresponding visible light capturing elements or light capturing elements that can be used to capture data, including images of scenes with unknown objects.

[0202] The head-mounted wearable device 116 includes a memory 1302 that stores instructions for performing a subset or all of the functions described herein. The memory 1302 may also include storage means.

[0203] like Figure 13As shown, the high-speed circuit system 1326 includes a high-speed processor 1330, a memory 1302, and a high-speed wireless circuit system 1332. In some examples, an image display driver 1320 is coupled to the high-speed circuit system 1326 and operated by the high-speed processor 1330 to drive the left and right image displays of the image display 1318 of the optical components. The high-speed processor 1330 can be any processor capable of managing the high-speed communication and operation of any general-purpose computing system required by the head-worn wearable device 136. The high-speed processor 1330 includes the processing resources required to manage high-speed data transmission over the high-speed wireless connection 1314 to a wireless local area network (WLAN) using the high-speed wireless circuit system 1332. In some examples, the high-speed processor 1330 executes the operating system of the head-worn wearable device 116 (e.g., a LINUX operating system) or other such operating system, and this operating system is stored in the memory 1302 for execution. Among other duties, the high-speed processor 1330, which executes the software architecture of the head-worn wearable device 116, manages data transmission with the high-speed wireless circuit system 1332. In some examples, the high-speed wireless circuit system 1332 is configured to implement the Institute of Electrical and Electronics Engineers (IEEE) 802.11 communication standard, also referred to herein as WiFi. In some examples, the high-speed wireless circuit system 1332 may implement other high-speed communication standards.

[0204] The low-power wireless circuit system 1334 and high-speed wireless circuit system 1332 of the head-mounted wearable device 136 may include a short-range transceiver (Bluetooth™) and a wireless wide area network transceiver, a wireless local area network transceiver, or a wide area network transceiver (e.g., cellular or WiFi). The mobile device 114, including transceivers communicating via low-power wireless connection 1312 and high-speed wireless connection 1314, can be implemented using details of the architecture of the head-mounted wearable device 116, such as other elements of network 1316.

[0205] Memory 1302 includes any storage device capable of storing various data and applications, including camera data generated by the left and right visible light cameras 1306, the infrared camera 1310, and the image processor 1322, as well as the generated images for display on an image display 1318 in the optical components by the image display driver 1320. While memory 1302 is shown as integrated with the high-speed circuitry 1326, in some examples, memory 1302 may be a separate, independent component of the head-mounted wearable device 116. In some such examples, electrical wiring may provide a connection from the image processor 1322 or the low-power processor 1336 to memory 1302 via a chip including the high-speed processor 1330. In some examples, the high-speed processor 1330 may manage addressing of memory 1302 such that the low-power processor 1336 will activate the high-speed processor 1330 whenever a read or write operation involving memory 1302 is required.

[0206] like Figure 13 As shown, the low-power processor 1336 or high-speed processor 1330 of the head-mounted wearable device 136 may be coupled to a camera device (visible light camera 1306, infrared emitter 1308 or infrared camera 1310), an image display driver 1320, a user input device 1328 (e.g., a touch sensor or a press button), and a memory 1302.

[0207] The head-mounted wearable device 116 is connected to a host computer. For example, the head-mounted wearable device 116 is paired with the mobile device 114 via a high-speed wireless connection 1314, or connected to the server system 1304 via a network 1316. The server system 1304 may be one or more computing devices as part of a service or network computing system, for example, it includes a processor, memory, and network communication interfaces to communicate with the mobile device 114 and the head-mounted wearable device 116 via the network 1316.

[0208] Mobile device 114 includes a processor and a network communication interface coupled to the processor. The network communication interface allows communication via network 1316, low-power wireless connection 1312, or high-speed wireless connection 1314. Mobile device 114 may also store in its memory at least a portion of instructions for generating binaural audio content to implement the functions described herein.

[0209] The output components of the head-mounted wearable device 116 include visual components, such as displays like LCDs, PDPs, LED displays, projectors, or waveguides. The image display of the optical components is driven by an image display driver 1320. The output components of the head-mounted wearable device 116 also include acoustic components (e.g., speakers), haptic components (e.g., vibration motors), other signal generators, etc. The input components (e.g., user input device 1328) of the head-mounted wearable device 116, mobile device 114, and server system 1304 may include alphanumeric input components (e.g., keyboards, touchscreens configured to receive alphanumeric input, photoelectric 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 that provide the position and force of a touch or touch gesture, or other haptic input components), audio input components (e.g., microphones), etc.

[0210] The head-mounted wearable device 116 may also include additional peripheral device elements. Such peripheral device elements may include biometric sensors, additional sensors, or display elements integrated with the head-mounted wearable device 116. For example, peripheral device elements may include any I / O components, including output components, motion components, position components, or any other such elements described herein.

[0211] For example, biometric components include those for detecting expressions (e.g., hand gestures, facial expressions, voice expressions, body posture, or eye tracking), measuring biosignals (e.g., blood pressure, heart rate, body temperature, sweating, or brainwaves), and identifying a person (e.g., voice recognition, retinal recognition, facial recognition, fingerprint recognition, or EEG-based recognition). Biometric components may include a BMI system that allows communication between the brain and external devices or machines. This can be achieved by recording brain activity data, converting that data into a format that can be understood by a computer, and then using the resulting signals to control devices or machines.

[0212] Motion components include accelerometer components (e.g., accelerometers), gravity sensor components, rotation sensor components (e.g., gyroscopes), etc. Position components include position sensor components (e.g., GPS receiver components) for generating position coordinates, Wi-Fi or Bluetooth™ transceivers for generating positioning system coordinates, altitude sensor components (e.g., altimeters or barometers that detect air pressure, from which altitude can be obtained), orientation sensor components (e.g., magnetometers), etc. Such positioning system coordinates can also be received from the mobile device 114 via low-power wireless circuitry system 1334 or high-speed wireless circuitry system 1332 through low-power wireless connection 1312 and high-speed wireless connection 1314.

[0213] Glossary

[0214] "Carrier signal" refers to any intangible medium, such as a medium capable of storing, encoding, or carrying instructions to be executed by a machine and including digital or analog communication signals, or other intangible medium that facilitates the communication of such instructions. Instructions can be transmitted or received over a network using a transmission medium via a network interface device.

[0215] "Client device" means 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, portable digital assistants (PDAs), smartphones, tablet computers, ultrabooks, netbooks, laptop computers, multiprocessor systems, microprocessor-based or programmable consumer electronics, game consoles, STBs, or any other communication device that a user can use to access the network.

[0216] "Communication network" refers to one or more parts of a network, such as an ad hoc network, intranet, extranet, virtual private network (VPN), local area network (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 GSM evolution (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 remote protocols, or other data transmission technologies.

[0217] A "component" refers to a logical or physical entity having boundaries defined by functional or subroutine calls, branch points, APIs, or other technologies that partition or modularize a particular processing or control function. A component can be combined with other components via its interface to perform machine processing. A component can be a packaged functional hardware unit designed for use with other components, and part of a program that typically performs a related function. A component can constitute a software component (e.g., code implemented on a machine-readable medium) or a hardware component.

[0218] A “hardware component” is a tangible unit capable of performing certain operations and can be configured or arranged in a physical manner. In various examples, one or more hardware components (e.g., processors or processor groups) of a computer system (e.g., a standalone computer system, a client computer system, or a server computer system) or a computer system can be configured by software (e.g., an application or an application portion) to operate to perform certain operations as described herein.

[0219] 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 application-specific integrated circuit (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 no longer a general-purpose processor. It should 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. Therefore, 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.

[0220] Consider an example where hardware components are temporarily configured (e.g., programmed), without requiring each hardware component to be configured or instantiated at any given time. For example, in cases where the hardware components include a general-purpose processor configured by software to become a dedicated processor, this general-purpose processor can be configured at different times as its own distinct dedicated processor (e.g., including different hardware components). The software accordingly configures one or more specific processors to constitute a particular hardware component at one time and different hardware components at different times. Hardware components can provide information to and receive information from other hardware components. Therefore, the described hardware components can be considered communicatively coupled. In cases where multiple hardware components exist simultaneously, communication can be achieved through signal transmission between or among two or more hardware components (e.g., via appropriate circuitry and buses). 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 the 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. Then, additional hardware components can access the memory device at a later time to retrieve and process the stored output. The hardware components can also initiate communication with input or output devices and can operate on resources (e.g., collections of information). The various operations of the example methods described herein can be performed, at least in part, by one or more processors configured, either temporarily (e.g., by software) or permanently, to perform the relevant operations. Whether temporarily or permanently configured, such processors can constitute components of a processor implementation that operates to perform one or more operations or functions described herein.

[0221] 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 in part by processors, where a particular processor or one or more processors are examples of hardware. For example, at least some of the various 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 related operations in a "cloud computing" environment or as a "Software as a Service" (SaaS) operation. 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 are accessible via a network (e.g., the Internet) and via one or more appropriate interfaces (e.g., APIs). The execution of some operations can be distributed among processors, not residing within a single machine, but deployed across multiple machines. In some examples, the processor or processor-implemented component may reside in a single geographic location (e.g., in a home environment, office environment, or server cluster). In other examples, the processor or processor-implemented component may be distributed across multiple geographic locations.

[0222] "Computer-readable storage medium" refers to both, for example, machine storage media and transmission media. Therefore, these terms include both storage devices / media and carrier / modulated data signals. The terms "machine-readable medium," "computer-readable medium," and "device-readable medium" refer to the same thing and can be used interchangeably in this disclosure. "Temporary message" refers to a message that is accessible for a limited time period, for example. A temporary message can be text, an image, video, etc. The access time of a temporary 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 technique, the message is temporary.

[0223] "Machine storage medium" refers to one or more storage devices and media (e.g., centralized or distributed databases, and associated caches and servers) that store executable instructions, routines, and data. Therefore, this term should be considered to include, but is not limited to, solid-state memory and optical and magnetic media, including memory internal or external to the processor. Specific examples of machine storage media, computer storage media, and device storage media include: non-volatile memory, including, for example, semiconductor memory devices such as erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), FPGAs, and flash memory devices; disks, such as internal hard disks and removable disks; magneto-optical disks; and CD-ROM and DVD-ROM disks. The terms "machine storage medium," "device storage medium," and "computer storage medium" refer to the same thing and may be used interchangeably in this disclosure.

[0224] The terms “machine storage medium,” “computer storage medium,” and “device storage medium” explicitly exclude carrier waves, modulated data signals, and other such media, at least some of which are encompassed within the term “signal medium.” “Non-transitory computer-readable storage medium” refers to, for example, a tangible medium capable of storing, encoding, or carrying instructions executable by a machine. “Signal medium” refers to, for example, any intangible medium capable of storing, encoding, or carrying instructions executable by a machine and includes digital or analog communication signals or other intangible media that facilitate the communication of software or data. The term “signal medium” should be considered to include any form of modulated data signal, carrier wave, etc. The term “modulated data signal” means a signal whose characteristics are set or altered in a manner that encodes information in the signal. The terms “transmission medium” and “signal medium” refer to the same thing and may be used interchangeably in this disclosure.

[0225] "User device" means, for example, a device that is accessed, controlled, or owned by a user and that the user interacts with to perform actions or interactions (including interactions with other users or computer systems). "Carrier signal" means any intangible medium or other intangible medium capable of storing, encoding, or carrying machine-executable instructions and including digital or analog communication signals. Instructions can be transmitted or received over a network using a transmission medium via a network interface device. "Client device" means 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, STBs, or any other communication device that a user can use to access the network.

[0226] "Communication network" refers to one or more parts of a network, which can be an ad hoc network, intranet, extranet, VPN, LAN, WLAN, WAN, WWAN, MAN, the Internet, a part of the Internet, a part of the PSTN, POTS network, cellular telephone network, wireless network, Wi-Fi® network, other types 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 CDMA connection, a GSM connection, or other types of cellular or wireless coupling. In this example, the coupling can implement any data transmission technology of various types, such as 1xRTT, EVDO technology, GPRS technology, EDGE technology, 3GPP including 3G, 4G networks, UMTS, HSPA, WiMAX, LTE standards, other data transmission technologies defined by various standards setting organizations, other telematics, or other data transmission technologies.

[0227] 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 described herein.

[0228] 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 an FPGA or 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 no longer a general-purpose processor. It should 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. Therefore, 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.

[0229] The various operations of the example methods described herein can be performed, at least in part, by one or more processors that are temporarily configured (e.g., by software) or permanently configured to perform the relevant operations. Whether temporarily or permanently configured, such processors can constitute the components of a processor implementation that performs the operations or functions described herein.

[0230] 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.

Claims

1. A method comprising: Access the monocular image of the depicted object; By applying a first machine learning model to the monocular image, both the volumetric reconstruction tensor of the monocular image and the pose of the object are predicted. The identifier corresponds to the pose of the object at a point in the canonical space associated with a set of location-encoded information; Obtain the point of the volume reconstruction tensor corresponding to the portion of the identified pose; By applying the second machine learning model together with the set of location encoding information to the points of the obtained volume reconstruction tensor, the obtained points are classified as being inside or outside the canonical volume; as well as In response to classifying the obtained points as inside or outside the canonical volume, a three-dimensional (3D) mesh representing the object is generated in the canonical space.

2. The method according to claim 1, wherein, The pose of the object includes the skeleton representation of the object.

3. The method according to any one of claims 1 to 2, further comprising: Select a point in the canonical space that is associated with the set of location encoding information.

4. The method according to any one of claims 1 to 3, further comprising: Select the second part of the pose of the object corresponding to the second point associated with the second set of position coding information in the canonical space; Obtain the second point of the volume reconstruction tensor corresponding to the second part of the pose; as well as By applying the second machine learning model together with a second set of location encoding information to a second point of the volume reconstruction tensor, the second point is classified as either inside or outside the canonical volume.

5. The method of claim 4, further comprising, for each portion of the normed volume, repeatedly selecting portions of the pose, and repeatedly classifying points of the volume reconstruction tensor corresponding to the selected portions of the pose as inside or outside the normed volume.

6. The method according to any one of claims 1 to 5, wherein, The set of location-coded information represents different body parts of the standard volume.

7. The method according to any one of claims 1 to 6, further comprising training the first machine learning model and the second machine learning model by performing a training operation, the training operation comprising: Obtain a set of training images depicting an object, each of the training images being associated with a 3D scan of the object depicted in the set of training images; as well as By modifying the pose of the 3D scan of the object to correspond to the pose in the canonical space, a ground-value canonical space representation of the object depicted in the set of training images is generated.

8. The method according to claim 7, wherein, The training operations include: Select a first training image depicting a first object, the first training image corresponding to a first truth-normalized space representation in the truth-normalized space representation of the first object; By applying the first machine learning model to the first training image, both the single volume reconstruction tensor of the monocular image and the single pose of the first object are predicted. The identifier represents a single portion of a single pose of the first object corresponding to a single point in the canonical space associated with a single set of location-encoded information; Obtain a single point corresponding to the single volume reconstruction tensor of the single part; By applying the second machine learning model along with a single set of location-encoded information to a single point of the volume reconstruction tensor, the single point is classified as either inside or outside a single canonical volume; and Occupancy loss is generated based on the deviation between the classification obtained by classifying the individual point and the corresponding part of the first truth-value canonical space representation of the object.

9. The method according to any one of claims 1 to 8, further comprising updating one or more parameters of the first machine learning model and the second machine learning model based on the occupancy loss.

10. The method according to any one of claims 1 to 9, further comprising: In response to classifying the individual point as inside or outside the canonical volume, a single 3D mesh representing the first object is generated in the canonical space; The single 3D mesh is set based on a single pose of the first object predicted by the first machine learning model; as well as The setting loss is calculated based on the deviation between the set single 3D mesh and the 3D scan of the first object.

11. The method according to any one of claims 1 to 10, further comprising: Cropping the portion of the monocular image corresponding to the head of the object; Based on the cropped portion of the monocular image, a facial landmark of the object's head is generated; Based on the generated facial landmarks, a head-specific canonical space reconstruction is predicted by applying a head-specific canonical network to the cropped portion. as well as The head-specific canonical space reconstruction is combined with the canonical volume corresponding to the object.

12. The method of claim 11, further comprising: Select a facial landmark in the head-specific canonical space representation of the object; Identify the specific portion of the object depicted in the cropped monocular image that corresponds to the selected facial landmark; Obtain a single point of the volume reconstruction tensor corresponding to the specific portion; By applying a third machine learning model to the obtained single point, the single point is classified as either inside or outside the head-specific canonical space representation of the object; and In response to classifying the obtained individual points as inside or outside the head-specific canonical space representation, the head-specific canonical space reconstruction is generated.

13. The method according to any one of claims 1 to 12, further comprising: By applying a fourth machine learning model to the cropped portion of the monocular image, both the additional volumetric reconstruction tensor of the cropped portion of the monocular image and the head pose of the object are predicted, wherein the individual point is obtained from the additional volumetric reconstruction tensor.

14. The method according to any one of claims 1 to 13, further comprising: The pose of the head region in the canonical volume is updated based on the facial landmarks, wherein the 3D mesh is generated using the canonical volume with the updated pose of the head region.

15. The method according to any one of claims 1 to 14, wherein, The canonical volume represents the full-body canonical space reconstruction of the object, and also includes: Identify the area where the whole-body canonical space reconstruction of the object overlaps with the head-specific canonical space reconstruction; and The region is smoothly deformed based on the difference between the whole-body canonical space reconstruction of the object and the head-specific canonical space reconstruction.

16. The method of claim 15, further comprising generating the 3D mesh based on the whole-body canonical space reconstruction, the head-specific canonical space reconstruction, and the smoothly deformed region.

17. The method according to any one of claims 1 to 16, further comprising generating an extended reality (XR) object based on the 3D mesh.

18. The method of claim 17, further comprising animates the XR object in a view of a real-world environment captured in real time by a camera device of a user system.

19. A system comprising: At least one processor; as well as At least one memory component having instructions stored thereon, the instructions causing the at least one processor to perform operations when executed by the at least one processor, the operations including: Access the monocular image of the depicted object; By applying a first machine learning model to the monocular image, both the volumetric reconstruction tensor of the monocular image and the pose of the object are predicted. The identifier corresponds to the pose of the object at a point in the canonical space associated with a set of location-encoded information; Obtain the point of the volume reconstruction tensor corresponding to the portion of the identified pose; By applying a second machine learning model along with the set of location-encoded information to the points of the obtained volume reconstruction tensor, the obtained points are classified as either inside or outside the canonical volume; and In response to classifying the obtained points as inside or outside the canonical volume, a three-dimensional (3D) mesh representing the object is generated in the canonical space.

20. A non-transitory computer-readable storage medium having instructions stored thereon, the instructions causing the at least one processor to perform an operation when executed by at least one processor, the operation comprising: Access the monocular image of the depicted object; By applying a first machine learning model to the monocular image, both the volumetric reconstruction tensor of the monocular image and the pose of the object are predicted. The identifier corresponds to the pose of the object at a point in the canonical space associated with a set of location-encoded information; Obtain the point of the volume reconstruction tensor corresponding to the portion of the identified pose; By applying the second machine learning model together with the set of location encoding information to the points of the obtained volume reconstruction tensor, the obtained points are classified as being inside or outside the canonical volume; as well as In response to classifying the obtained points as inside or outside the canonical volume, a three-dimensional (3D) mesh representing the object is generated in the canonical space.

Images