Neural network control variables
Directly mapping the origin and direction of the light ray to the pixel value through the neural light field network, solving the problem of high resource consumption in the NeRF network, improving the speed and efficiency of 3D scene generation, suitable for devices with limited resources, and improving user experience.
Patent Information
- Application Number
- CN202380031568.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2022-03-28
- Filing Date
- 2023-03-28
- Publication Date
- 2024-11-15
AI Technical Summary
The existing NeRF network consumes a lot of resources and operates in low efficiency when generating 3D scenes, resulting in a decrease in user experience, especially on devices with limited resources.
Using a neural light field (NeLF) network, by directly mapping the origin and direction of the light ray to RGB or pixel values, sampling at multiple points along the light ray is avoided, and optimized on the basis of the NeRF network to reduce resource consumption.
Improves the speed and efficiency of generating 3D scenes, is suitable for resource-constrained devices, improves user experience, and reduces the total amount of system resources.
Smart Images

Figure CN118974777A_ABST
Abstract
Description
[0001] Priority claim
[0002] This application claims the benefit of priority to U.S. patent application serial number 17 / 656,778, filed on March 28, 2022, the entire contents of which are incorporated herein by reference. Technical Field
[0003] The present disclosure generally relates to providing 3D modeling of real-world environments using messaging applications. Background Art
[0004] Three-dimensional (3D) reconstruction from multiple two-dimensional (2D) images is the creation of a 3D model from a set of images. The process typically involves receiving 2D images from multiple viewpoints and then reconstructing a 3D model based on these images. The 3D model can be used to simulate a 3D view of a real-world environment to produce different experiences such as augmented reality experiences and virtual reality experiences. BRIEF DESCRIPTION OF THE DRAWINGS
[0005] In the drawings, which are not necessarily drawn to scale, similar reference numerals may describe similar components in different views. To easily identify the discussion of any particular element or action, the highest digit or digits in the reference numeral refer to the figure number in which the element is first introduced. Some non-limiting examples are shown in the accompanying drawings, in which:
[0006] Figure 1 is a diagrammatic representation of a networking environment in which the present disclosure may be deployed, according to some examples.
[0007] Figure 2 is a diagrammatic representation of a messaging client application according to some examples.
[0008] Figure 3 is a diagrammatic representation of data structures as maintained in a database according to some examples.
[0009] Figure 4 is a graphical representation of messages according to some examples.
[0010] Figure 5 is a block diagram illustrating an example neural light field (NLF) estimation system according to some examples.
[0011] Figure 6 and Figure 7 is a diagrammatic representation of components of an NLF system according to some examples.
[0012] Figure 8 is a diagrammatic representation of components of an example neural radiance field (NRF) network system according to some examples.
[0013] Fig. 9 is a flow chart illustrating example operation of a NLF estimation system according to some examples.
[0014] Fig.10 is a diagrammatic representation of a machine in the form of a computer system according to some examples within which a set of instructions may be executed for causing the machine to perform any one or more of the methodologies discussed herein.
[0015] Fig.11 is a block diagram illustrating a software architecture in which examples may be implemented. DETAILED DESCRIPTION
[0016] The following description includes systems, methods, techniques, instruction sequences, and computing machine program products that embody illustrative examples of the present disclosure. In the following description, for the purpose of illustration, many specific details are set forth to provide an understanding of various examples. However, it will be apparent to those skilled in the art that examples may be practiced without these specific details. Typically, known instruction instances, protocols, structures, and techniques are not necessarily shown in detail.
[0017] Inferring representations of 3D scenes from 2D observations is a fundamental problem in computer graphics and computer vision. Neural Radiance Field (NeRF) networks, trained from simple multi-layer perceptron (MLP) networks, are one way to estimate or predict some of the 2D observations needed to generate 3D scenes. However, in order to perform the necessary predictions, NeRF networks consume a lot of resources and operate inefficiently. This results in delayed rendering times for even a single pixel, slowing down the overall generation of 3D scenes and reducing the overall user experience and enjoyment of the system. Specifically, NeRF networks aggregate the radiance of hundreds of sample points along a ray via alpha synthesis and use hundreds of network forwards. This makes such networks very slow, especially on resource-constrained devices such as mobile devices.
[0018] One way to address these shortcomings is to reduce the model size of the NeRF MLP network. However, this results in a severe degradation in the quality of the rendered image while only reducing the time by a minimal amount. Other approaches to address these deficiencies involve reducing the number of sampling points along the rays. However, these approaches still involve multiple iterations and network forwards, which still consume a lot of system resources.
[0019] The disclosed technology improves the efficiency of electronic devices by providing a machine learning technology (model) that can compactly represent 3D scenes based on 2D image inputs. The disclosed machine learning model includes a neural light field (NeLF or NLF) instead of a NeRF network. The NeLF network can directly map the origin and direction of light to the associated RGB or pixel value. The disclosed technology avoids having to sample multiple points along the camera light and process each such point separately with a NeRF network to estimate radiation. This enhances the speed and efficiency with which 3D scenes can be generated and used to provide various end-user experiences such as augmented reality (AR) and / or virtual reality (VR) experiences. That is, the NeRF technology involves reapplying the NeRF network to each point along the array and then combining the radiation values to estimate the pixel value, while the disclosed technology can estimate the pixel value through a single forward pass of the NeLF network. This makes the disclosed network suitable for resource-constrained devices because the amount of resources required to complete the task is reduced.
[0020] Specifically, the disclosed technology receives a set of 2D images representing a first view of a real-world environment. The disclosed technology applies a machine learning model including a neural light field network to the set of 2D images to predict pixel values of a target image representing a second view of the real-world environment, and the machine learning model is trained to directly map the origin and direction of light to a given pixel value. The disclosed technology generates a 3D model of the real-world environment based on the set of 2D images and the predicted target image. This improves the overall experience of the user using the electronic device and reduces the total amount of system resources required to complete the task.
[0021] In some examples, in order to improve or speed up the training of the disclosed NeLF machine learning model (or machine learning technique), the disclosed examples use a NeRF network to generate a training data set that includes pseudo-generated training images. The pseudo-generated training images include different camera poses of a real-world environment. By using a NeRF network to provide training data (e.g., as a teacher network), a reduced number of real images of a given real-world environment need to be captured. Additional images are generated using a small number of real images using the NeRF network. These additional images will then be used to train the NeLF network to estimate or predict pixel values based on new or selected camera poses. That is, pseudo data is presented from random views of a real-world environment using a pre-trained NeRF network or model.
[0022] Networked computing environment
[0023] Figure 11 is a block diagram illustrating an example messaging system 100 for exchanging data (e.g., messages and associated content) over a network. The messaging system 100 includes multiple instances of a client device 102, each instance hosting several applications, including a messaging client 104 and other external applications 109 (e.g., third-party applications). Each messaging client 104 is communicatively coupled to other instances of the messaging client 104 (e.g., hosted on corresponding other client devices 102), a messaging server system 108, and an external APP server 110 via a network 112 (e.g., the Internet). The messaging client 104 can also communicate with locally hosted third-party applications (e.g., external app 109) using an application programming interface (API).
[0024] The client device 102 machine 1100 can operate as a standalone device, or can be coupled (e.g., networked) to other machines. In a networked deployment, the client device 102 can operate with the performance of a server machine or a client machine in a server-client network environment, or operate as a peer machine in a peer-to-peer (or distributed) network environment. The client device 102 may include, but is not limited to: a server computer, a client computer, a personal computer (PC), a tablet computer, a laptop computer, a netbook, a set-top box (STB), a personal digital assistant (PDA), an entertainment media system, a cellular phone, a smart phone, a mobile device, a wearable device (e.g., a smart watch), a smart home device (e.g., a smart appliance), other smart devices, a web device, a network router, a network switch, a network bridge, or any machine capable of performing the disclosed operations. In addition, although only a single client device 102 is shown, the term "client device" should also be considered to include a collection of machines that perform the disclosed operations individually or jointly.
[0025] In some examples, client device 102 may include AR glasses or an AR head-mounted device, wherein virtual content or elements are displayed within the lenses of the glasses while the user views the real world environment through the lenses. For example, images may be presented on a transparent display to allow the user to simultaneously view virtual content presented on the display and objects in the real world.
[0026] The messaging clients 104 are able to communicate and exchange data with other messaging clients 104 and messaging server systems 108 via the network 112. The data exchanged between the messaging clients 104 and between the messaging clients 104 and messaging server systems 108 include functions (e.g., commands to activate functions) and payload data (e.g., text, audio, video, or other multimedia data).
[0027] The messaging server system 108 provides server-side functionality to specific messaging clients 104 via the network 112. Although certain functions of the messaging system 100 are described herein as being performed by the messaging client 104 or by the messaging server system 108, the location of certain functions within the messaging client 104 or within the messaging server system 108 may be a design choice. For example, it may be technically preferred to initially deploy certain technologies and functions within the messaging server system 108, but later migrate the technologies and functions to the messaging client 104 where the client device 102 has sufficient processing power.
[0028] The messaging server system 108 supports various services and operations provided to the messaging clients 104. Such operations include sending data to the messaging clients 104, receiving data from the messaging clients 104, and processing data generated by the messaging clients 104. As examples, the data may include message content, client device information, geographic location information, media enhancements and overlays, message content persistence conditions, social network information, and live event information. The data exchange within the messaging system 100 is activated and controlled by functions available via a user interface (UI) of the messaging clients 104.
[0029] Turning now specifically to the messaging server system 108, an API server 116 is coupled to and provides a programming interface to the application server 114. The application server 114 is communicatively coupled to a database server 120, which facilitates access to a database 126 that stores data associated with messages processed by the application server 114. Similarly, a web server 128 is coupled to and provides a web-based interface to the application server 114. To this end, the web server 128 processes incoming network requests via the Hypertext Transfer Protocol (HTTP) and several other related protocols.
[0030] The API server 116 receives and sends message data (e.g., commands and message payloads) between the client device 102 and the application server 114. Specifically, the application API server 116 provides a set of interfaces (e.g., routines and protocols) that can be called or queried by the messaging client 104 to activate the functionality of the application server 114. The API server 116 exposes various functions supported by the application server 114, including: account registration; login functionality; sending messages from a particular messaging client 104 to another messaging client 104 via the application server 114; sending media files (e.g., images or videos) from the messaging client 104 to the messaging server 118 and for possible access by another messaging client 104; setting of media data collections (e.g., stories); retrieving a friend list of the user of the client device 102; retrieving such collections; retrieving messages and content; adding and removing entities (e.g., friends) from an entity graph (e.g., a social graph); locating friends in a social graph; and opening application events (e.g., related to the messaging client 104).
[0031] The application server 114 hosts several server applications and subsystems, including, for example, a messaging server 118, an image processing server 122, and a social network server 124. The messaging server 118 implements several message processing technologies and functions, particularly those related to the aggregation and other processing of content (e.g., text and multimedia content) included in messages received from multiple instances of the messaging client 104. As will be described in further detail, text and media content from multiple sources can be aggregated into collections of content (e.g., called stories or galleries). These collections are then made available to the messaging client 104. In view of the hardware requirements of other processor- and memory-intensive data processing, such processing can also be performed on the server side by the messaging server 118.
[0032] The application server 114 includes an image processing server 122 that is dedicated to performing various image processing operations, typically with respect to images or videos within the payload of messages sent from or received at the messaging server 118 .
[0033] The image processing server 122 is used to implement the scanning function of the enhancement system 208 (such as Figure 2). The scanning function includes activating and providing one or more AR (or VR) experiences on the client device 102 when an image or a set of 2D images is captured by the client device 102. Specifically, the messaging application 104 on the client device 102 can be used to activate the camera. The camera displays one or more real-time 2D images or 2D videos together with one or more icons or identifiers of one or more AR experiences to the user. The user can select a given identifier in the identifier to start the corresponding AR experience or perform the desired image modification.
[0034] The social network server 124 supports various social networking functions and services and makes them available to the messaging server 118. To this end, the social network server 124 maintains and accesses an entity graph 308 (e.g., Figure 3 ). Examples of functions and services supported by the social networking server 124 include identifying other users of the messaging system 100 that have a relationship with a particular user or that the particular user is "following," and identifying interests and other entities of a particular user.
[0035] Returning to the messaging client 104, the features and functions of the external resource (e.g., a third-party application 109 or an applet) are available to the user via the interface of the messaging client 104. The messaging client 104 receives a user selection of an option for launching or accessing features of an external resource (e.g., a third-party resource) such as an external application 109. The external resource can be a third-party application (external application 109) installed on the client device 102 (e.g., a "local application"), or a small-scale version of a third-party application (e.g., a "applet") hosted on the client device 102 or away from the client device 102 (e.g., on a third-party server 110). The small-scale version of the third-party application includes a subset of the features and functions of the third-party application (e.g., a full-scale local version of a third-party standalone application) and is implemented using a markup language document. In one example, the small-scale version of the third-party application (e.g., a "applet") is a web-based markup language version of the third-party application and is embedded in the messaging client 104. In addition to using markup language documents (eg, *ml files), applets may include scripting languages (eg, .*js files or .json files) and style sheets (eg, *ss files).
[0036] In response to receiving a user selection of an option for launching or accessing a feature of an external resource (external application 109), the messaging client 104 determines whether the selected external resource is a web-based external resource or a locally installed external application. In some cases, the external application 109 installed locally on the client device 102 can be launched independently of the messaging client 104 and separately from the messaging client 104, for example, by selecting an icon corresponding to the external application 109 on the home screen of the client device 102. A small-scale version of such an external application can be launched or accessed via the messaging client 104, and in some examples, no part of the small-scale external application can be (or only a limited part can be) accessed outside the messaging client 104. The small-scale external application can be launched by the messaging client 104 receiving a markup language document associated with the small-scale external application from the external APP server 110 and processing such a document.
[0037] In response to determining that the external resource is a locally installed external application 109, the messaging client 104 instructs the client device 102 to launch the external application 109 by executing the locally stored code corresponding to the external application 109. In response to determining that the external resource is a web-based resource, the messaging client 104 communicates with the external APP server 110 to obtain a markup language document corresponding to the selected resource. The messaging client 104 then processes the obtained markup language document to present the web-based external resource within the user interface of the messaging client 104.
[0038] The messaging client 104 may notify the user of the client device 102 or other users (e.g., "friends") associated with such a user of activities occurring in one or more external resources. For example, the messaging client 104 may provide a notification to a participant in a conversation (e.g., a chat session) in the messaging client 104 about the current or recent use of an external resource by one or more members of a group of users. One or more users may be invited to join an active external resource or to start an external resource that was recently used but is currently inactive (in the friend group). The external resource may provide the participants in the conversation (each participant using a corresponding messaging client 104) with the ability to share items, conditions, states, or locations in the external resource with one or more members of the user group in the chat session. The shared item may be an interactive chat card that members of the chat may interact with, for example, to start a corresponding external resource, view specific information within an external resource, or bring members of the chat to a specific location or state within an external resource. Within a given external resource, a response message may be sent to a user on the messaging client 104. The external resource may selectively include different media items in the response based on the current context of the external resource.
[0039] The messaging client 104 may present a list of available external resources (e.g., third-party or external applications 109 or applets) to the user to launch or access a given external resource. The list may be presented in a context-sensitive menu. For example, icons representing different external applications 109 (or applets) may vary based on how the user launches the menu (e.g., from a conversational interface or from a non-conversational interface).
[0040] In some examples, the messaging client 104 may generate a 3D model of the real world environment. To this end, the messaging client 104 may implement the NLF estimation system 224 ( Figure 2 ). The NLF estimation system 224 includes a machine learning model that can receive one or more 2D images associated with a first view of a real-world environment and can directly predict one or more pixel values of an image associated with a second view of the real-world environment. The received one or more 2D images can be processed together with the predicted one or more pixel values of the image to generate a 3D model, for example by performing a 3D reconstruction process.
[0041] System Architecture
[0042] Figure 21 is a block diagram showing additional details about the messaging system 100 according to some examples. Specifically, the messaging system 100 is shown to include a messaging client 104 and an application server 114. The messaging system 100 contains several subsystems that are supported on the client side by the messaging client 104 and on the server side by the application server 114. These subsystems include, for example, a transient timer system 202, a collection management system 204, an enhancement system 208, a mapping system 210, a gaming system 212, and an NFL estimation system 224.
[0043] The transient timer system 202 is responsible for implementing temporary or time-limited access to content by the messaging client 104 and the messaging server 118. The transient timer system 202 includes several timers that selectively enable access (e.g., for presentation and display) to messages and associated content via the messaging client 104 based on duration and display parameters associated with a message or a collection of messages (e.g., a story). Additional details regarding the operation of the transient timer system 202 are provided below.
[0044] The collection management system 204 is responsible for managing collections or collections of media (e.g., collections of text, images, video, and audio data). Collections of content (e.g., messages, including images, videos, text, and audio) can be organized into "event libraries" or "event stories." Such collections can be made available for a specified time period (e.g., the duration of an event related to the content). For example, content related to a concert can be made available as a "story" for the duration of the concert. The collection management system 204 can also be responsible for publishing an icon that provides notification of the existence of a particular collection to the user interface of the messaging client 104.
[0045] In addition, the collection management system 204 also includes a curation interface 206, which enables collection managers to manage and curate specific content collections. For example, the curation interface 206 enables event organizers to curate collections of content related to specific events (e.g., deleting inappropriate content or redundant messages). In addition, the collection management system 204 uses machine vision (or image recognition technology) and content rules to automatically curate content collections. In some examples, compensation can be paid to users for including user-generated content in a collection. In such a case, the collection management system 204 operates to automatically pay such users for using their content.
[0046] The enhancement system 208 provides various functions that enable users to enhance (e.g., annotate or otherwise modify or edit) media content associated with a message. For example, the enhancement system 208 provides functions related to generating and publishing media overlays (media overlays) for messages processed by the messaging system 100. The enhancement system 208 is operable to provide media overlays or enhancements (e.g., image filters) to the messaging client 104 based on the geographic location of the client device 102. In another example, the enhancement system 208 is operable to provide media overlays to the messaging client 104 based on other information such as social network information of the user of the client device 102. Media overlays can include audio and visual content and visual effects. Examples of audio and visual content include pictures, text, logos, animations, and sound effects. Examples of visual effects include color overlays. Audio and visual content or visual effects can be applied to media content items (e.g., photos) at the client device 102. For example, media overlays can include text, graphic elements, or images that can be superimposed on the top of photos taken by the client device 102. In another example, the media overlay includes a location identification overlay (e.g., Venice Beach), the name of a live event, or a business name overlay (e.g., Beach Cafe). In another example, the augmentation system 208 uses the geographic location of the client device 102 to identify a media overlay that includes the name of a business at the geographic location of the client device 102. The media overlay may include other logos associated with the business. The media overlay may be stored in the database 126 and accessed through the database server 120.
[0047] In some examples, the enhancement system 208 provides a user-based publishing platform that enables a user to select a geographic location on a map and upload content associated with the selected geographic location. The user can also specify situations in which a particular media overlay should be provided to other users. The enhancement system 208 generates a media overlay that includes the uploaded content and associates the uploaded content with the selected geographic location.
[0048] In other examples, the augmentation system 208 provides a merchant-based publishing platform that enables merchants to select specific media overlays associated with a geographic location via a bidding process. For example, the augmentation system 208 associates the media overlay of the highest bidding merchant with the corresponding geographic location for a predefined amount of time. The augmentation system 208 communicates with the image processing server 122 to obtain an AR experience and presents an identifier of such an experience in one or more user interfaces (e.g., as an icon on a real-time image or video, or as a thumbnail or icon in an interface dedicated to the identifier of the presented AR experience). Once the AR experience is selected, one or more images, videos, or AR graphical elements are retrieved and presented as an overlay on the image or video captured by the client device 102. In some cases, the camera is switched to a front view (e.g., the front camera of the client device 102 is activated in response to the activation of a particular AR experience), and an image from the front camera of the client device 102 rather than the rear camera of the client device 102 begins to be displayed on the client device 102. One or more images, videos, or AR graphical elements are retrieved and presented as an overlay on top of the image captured and displayed by the front-facing camera of the client device 102 .
[0049] In other examples, the augmentation system 208 can communicate and exchange data with another augmentation system 208 on another client device 102 and a server via the network 112. The exchanged data may include: a session identifier identifying the shared AR session; a transformation between the first client device 102 and the second client device 102 (e.g., the plurality of client devices 102 include the first device and the second device), the transformation used to align the shared AR session to a common origin; a common coordinate system; functionality (e.g., a command to activate a function), and other payload data (e.g., text, audio, video, or other multimedia data).
[0050] The augmentation system 208 sends the transformation to the second client device 102 so that the second client device 102 can adjust the AR coordinate system based on the transformation. In this way, the first client device 102 and the second client device 102 synchronize their coordinate systems and frames to display the content in the AR session. Specifically, the augmentation system 208 calculates the origin of the second client device 102 in the coordinate system of the first client device 102. The augmentation system 208 can then determine an offset in the coordinate system of the second client device 102 based on the position of the origin in the coordinate system of the second client device 102 from the perspective of the second client device 102. The transformation is generated using the offset so that the second client device 102 generates AR content according to a common coordinate system with the first client device 102.
[0051] The augmentation system 208 can communicate with the client device 102 to establish a separate or shared AR session. The augmentation system 208 can also be coupled to the messaging server 118 to establish an electronic group communication session (e.g., group chat, instant messaging) for the client device 102 in the shared AR session. The electronic group communication session can be associated with a session identifier provided by the client device 102 to obtain access to the electronic group communication session and the shared AR session. In some examples, the client device 102 first obtains access to the electronic group communication session, and then obtains a session identifier in the electronic group communication session that allows the client device 102 to access the shared AR session. In some examples, the client device 102 is able to access the shared AR session without the assistance of the augmentation system 208 in the application server 114 or communication with the augmentation system 208 in the application server 114.
[0052] In some examples, the augmentation system 208 communicates with the NLF estimation system 224 to obtain a 3D model of the real-world environment. The augmentation system 208 may present a display of one or more virtual elements in a video based on the 3D model of the real-world environment, the video including one or more 2D images. That is, the augmentation system 208 may receive a selection of an AR or VR experience. The augmentation system 208 may retrieve one or more virtual elements associated with the selected AR / VR experience. The augmentation system 208 may display the virtual elements together with the real-world objects depicted in the 2D image based on the 3D model of the real-world environment to provide an AR or VR experience.
[0053] The mapping system 210 provides various geo-location functions and supports presentation of map-based media content and messages by the messaging client 104. For example, the mapping system 210 enables display of (e.g., stored in Figure 3 316) to indicate the current or past locations of the user's "friends," as well as media content (e.g., a collection of messages including photos and videos) generated by such friends within the context of the map. For example, messages posted by a user to the messaging system 100 from a particular geographic location may be displayed to the particular user's "friends" on the map interface of the messaging client 104 within the context of that particular location on the map. The user may also share his or her location and status information with other users of the messaging system 100 (e.g., using an appropriate status avatar) via the messaging client 104, where the location and status information is similarly displayed to selected users within the context of the map interface of the messaging client 104.
[0054] The gaming system 212 provides various gaming functions within the context of the messaging client 104. The messaging client 104 provides a gaming interface that provides a list of available games (e.g., web-based games or web-based applications) that can be launched by a user within the context of the messaging client 104 and played with other users of the messaging system 100. The messaging system 100 also enables a particular user to invite other users to participate in playing a particular game by sending an invitation to such other users from the messaging client 104. The messaging client 104 also supports both voice and text messaging (e.g., chat) within the context of gaming, provides leaderboards for gaming, and also supports providing in-game rewards (e.g., gaming coins and items).
[0055] The external resource system 220 provides the messaging client 104 with an interface for communicating with the external APP server 110 to launch or access external resources. Each external resource (application) server 110 hosts, for example, a small-scale version of an application based on a markup language (e.g., HTML5) or an external application (e.g., a game, utility, payment, or ride-sharing application outside the messaging client 104). The messaging client 104 can launch the web-based resource by accessing an HTML5 file from an external resource (APP) server 110 associated with a web-based resource (e.g., application). In some examples, the application hosted by the external resource server 110 is programmed in JavaScript using a software development kit (SDK) provided by the messaging server 118. The SDK includes APIs that have functions that can be called or activated by web-based applications. In some examples, the messaging server 118 includes a JavaScript library that provides a given third-party resource with access rights to certain user data of the messaging client 104. HTML5 is used as an example technology for programming games, but applications and resources programmed based on other technologies can be used.
[0056] In order to integrate the functionality of the SDK into the web-based resource, the SDK is downloaded from the messaging server 118 by the external resource (APP) server 110 or received by the external resource (app) server 110 in other ways. Once downloaded or received, the SDK will be 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 messaging client 104 into the web-based resource.
[0057] The SDK stored on the messaging server 118 effectively provides a bridge between external resources (e.g., a third party or external application 109 or applet and the messaging client 104). This provides users with a seamless experience of communicating with other users on the messaging client 104 while retaining the look and feel of the messaging client 104. In order to bridge the communication between the external resources and the messaging client 104, in some examples, the SDK facilitates the communication between the external resource server 110 and the messaging client 104. In some examples, the WebViewJavaScriptBridge running on the client device 102 establishes two one-way communication channels between the external resources and the messaging client 104. Messages are sent asynchronously via these communication channels between the external resources and the messaging client 104. Each SDK function activation is sent as a message and a callback. Each SDK function is implemented by constructing a unique callback identifier and sending a message with the callback identifier.
[0058] By using the SDK, not all information from the messaging client 104 is shared with the external resource server 110. The SDK limits which information is shared based on the needs of the external resource. In some examples, each external resource server 110 provides an HTML5 file corresponding to a web-based external resource to the messaging server 118. The messaging server 118 can add a visual representation (e.g., box art or other graphics) of the web-based external resource in the messaging client 104. Once the user selects the visual representation or instructs the messaging client 104 to access a feature of a web-based external resource through the graphical user interface (GUI) of the messaging client 104, the messaging client 104 obtains the HTML5 file and instantiates the resources required to access the features of the web-based external resource.
[0059] The messaging 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, the messaging client 104 determines whether the launched external resource has been previously authorized to access the user data of the messaging client 104. In response to determining that the launched external resource has been previously authorized to access the user data of the messaging client 104, the messaging client 104 presents another GUI of the external resource including the functions and features of the external resource. In response to determining that the launched external resource has not been previously authorized to access the user data of the messaging client 104, after a threshold time period (e.g., 3 seconds) of displaying the login page or title screen of the external resource, the messaging client 104 slides up a menu (e.g., animating the menu to emerge from the bottom of the screen to the middle or other portion of the screen) for authorizing the external resource to access the user data. The menu identifies the type of user data that the external resource will be authorized to use. In response to receiving the user selection of the accept option, the messaging client 104 adds the external resource to the list of authorized external resources and enables the external resource to access the user data from the messaging client 104. In some examples, the messaging client 104 authorizes the external resource to access the user data according to the OAuth 2 framework.
[0060] The messaging client 104 controls the type of user data shared with the external resource based on the type of external resource that is authorized. For example, an external resource including a full-scale external application (e.g., a third party or external application 109) is provided with access to a first type of user data (e.g., only a two-dimensional (2D) avatar of a user with or without different avatar features). As another example, an external resource including a small-scale version of an external application (e.g., a web-based version of a third-party application) is provided with access to a second type of user data (e.g., payment information, a 2D avatar of a user, a three-dimensional (3D) avatar of a user, and an avatar with various avatar features). Avatar representation characteristics include different ways to customize the look and feel of an avatar representation (e.g., different poses, facial features, clothing, etc.).
[0061] The NLF estimation system 224 includes a trained machine learning model that can receive one or more 2D images associated with a first view of a real-world environment and can generate or synthesize a new image of the real-world environment corresponding to a second view of the real-world environment. The NLF estimation system 224 predicts pixel values of the new image based on the one or more 2D images associated with the first view of the real-world environment. The NLF estimation system 224 can then perform 3D image reconstruction using the actually captured 2D images and the synthesized or predicted images to generate a 3D model of the real-world environment.
[0062] In some examples, the machine learning model of the NLF estimation system 224 can include a neural light field network (e.g., a deep residual MLP) that has a single forward pass for each camera ray associated with a given pixel of the new image. The machine learning model of the NLF estimation system 224 can be trained based on synthetic images or pseudo images generated by a different machine learning model (e.g., a machine learning model that implements a neural radiance field network that performs multiple forward passes for each ray to generate pixel values).
[0063] Data Architecture
[0064] Figure 3 is a diagram illustrating a data structure 300 that may be stored in a database 126 of a messaging server system 108 according to some examples. Although the contents of the database 126 are illustrated as including several tables, it should be understood that data may be stored in other types of data structures (eg, object-oriented databases).
[0065] Database 126 includes message data stored in message table 302. For any particular message, the message data includes at least message sender data, message recipient (or receiver) data, and payload. Figure 4 Additional details regarding information that may be included in a message and included in the message data stored in message table 302 are described.
[0066] The entity table 306 stores entity data and is linked (e.g., referenced) to the entity graph 308 and profile data 316. Entities for which records are maintained within the entity table 306 may include individuals, corporate entities, organizations, objects, places, events, etc. Regardless of the entity type, any entity about which the messaging server system 108 stores data may be an identified entity. Each entity is provided with a unique identifier as well as an entity type identifier (not shown).
[0067] The entity graph 308 stores information about relationships and associations between entities. Such relationships may be social, professional (e.g., working in a common company or organization), interest-based, or activity-based, by way of example only.
[0068] Profile data 316 stores multiple types of profile data about a particular entity. Based on the privacy settings specified by the particular entity, profile data 316 can be selectively used and presented to other users of the messaging system 100. In the case where the entity is a person, profile data 316 includes, for example, a user name, phone number, address, settings (e.g., notification and privacy settings), and an avatar representation (or a collection of such avatar representations) selected by the user. The particular user can then selectively include one or more of these avatar representations in the content of messages transmitted via the messaging system 100 and in a map interface displayed to other users by the messaging client 104. The collection of avatar representations can include a "status avatar" that presents a graphical representation of a state or activity that a user can select to communicate at a particular time.
[0069] Where the entity is a group, the group's profile data 316 may similarly include one or more visual representations associated with the group, in addition to the group name, members, and various settings related to the group (eg, notifications).
[0070] Database 126 also stores enhancement data, such as overlays or filters, in enhancement table 310. Enhancement data is associated with and applied to videos (video data is stored in video table 304) and images (image data is stored in image table 312).
[0071] The database 126 may also store data related to individual and shared AR sessions. The data may include data transmitted between an AR session client controller of a first client device 102 and another AR session client controller of a second client device 102, and data transmitted between an AR session client controller and the augmented system 208. The data may include data for establishing a common coordinate system for a shared AR scene, transformations between devices, session identifiers, images depicting the body, skeletal joint positions, wrist joint positions, feet, and the like.
[0072] In one example, a filter is an overlay displayed as an overlay on an image or video during presentation to a recipient user. Filters can be of various types, including filters selected by a user from a set of filters presented to a sending user by messaging client 104 when the sending user is composing a message. Other types of filters include geolocation filters (also referred to as geofilters), which can be presented to a sending user based on a geographic location. For example, a geolocation filter specific to a nearby or special location can be presented by messaging client 104 within a user interface based on geographic location information determined by a global positioning system (GPS) unit of client device 102.
[0073] Another type of filter is a data filter, which can be selectively presented to the sending user by the messaging client 104 based on other input or information collected by the client device 102 during the message creation process. Examples of data filters include the current temperature at a particular location, the current speed at which the sending user is traveling, the battery life of the client device 102, or the current time.
[0074] Other augmented data that may be stored in the image table 312 include AR content items (e.g., including one or more virtual elements corresponding to the applied AR experience). AR content items or AR items may be real-time special effects (e.g., including virtual graphic elements such as videos, images, or animations) and sounds that may be added to an image or video.
[0075] As described above, augmented data includes AR content items, overlays, image transformations, AR images, AR logos or badges, and similar items that refer to modifications that can be applied to image data (e.g., videos or images). This includes real-time modifications that modify images when they are captured using a device sensor (e.g., one or more cameras) of the client device 102 and then display the modified images on the screen of the client device 102. This also includes modifications to stored content (e.g., video clips in a library that can be modified). For example, in a client device 102 that can access multiple AR content items, a user can use a single video clip with multiple AR content items to see how different AR content items will modify the stored clips. For example, by selecting different AR content items for the content, multiple AR content items that apply different pseudo-random movement models can be applied to the same content. Similarly, real-time video capture can be used with the modifications shown to show how the video image currently captured by the sensor of the client device 102 will modify the captured data. Such data can be simply displayed on the screen without being stored in the memory, or the content captured by the device sensor can be recorded and stored in the memory with or without modification (or both). In some systems, the preview function can show different AR content items will be displayed in different windows of the display at the same time. This can, for example, enable viewing multiple windows with different pseudo-random animations on the display at the same time.
[0076] Thus, data and various systems using AR content items or other such transformation systems that modify content using the data may involve detection of objects (e.g., faces, hands, bodies, cats, dogs, surfaces, objects, etc.) in video frames; tracking of these objects as they leave, enter, and move around the field of view; and modification or transformation of these objects as they are tracked. In various examples, different methods for implementing such transformations may be used. Some examples may involve: generating a 3D mesh model of one or more objects; and using transformations and animated textures of the model within the video to implement the transformation. In other examples, tracking of points on the object may be used to place an image or texture (which may be 2D or 3D) at the tracked location. In further examples, neural network analysis of video frames may be used to place images, models, or textures in content (e.g., images or video frames). Thus, AR content items refer to both images, models, and textures used to create transformations in the content, as well as additional modeling and analysis information required to implement such transformations using object detection, tracking, and placement.
[0077] Real-time video processing can be performed using any kind of video data (e.g., video streams, video files, etc.) stored in the memory of any kind of computerized system. For example, a user can load video files and store them in the memory of the device, or a sensor of the device can be used to generate a video stream. In addition, any object, such as a human face and parts of a human body, an animal, or a non-living thing (e.g., a chair, a car, or other objects) can be processed using computer animation models.
[0078] In some examples, in the case where a specific modification is selected together with the content to be transformed, the element to be transformed is identified by a computing device, and then if the element to be transformed is present in a video frame, the element to be transformed is detected and tracked. The elements of the object are modified according to the request for modification, thereby transforming the frame of the video stream. The transformation of the frame of the video stream can be performed by different methods for different types of transformations. For example, for a frame transformation that mainly refers to a variation of the elements of the object, the feature points of each element of the object are calculated (for example, using an active shape model (ASM) or other known methods). Then, for each element of at least one element of the object, a grid based on the feature points is generated. The grid is used in a subsequent stage, in which the elements of the object in the video stream are tracked. During the tracking process, the grid mentioned for each element is aligned with the position of each element. Then, additional points are generated on the grid. A first group of first points is generated for each element based on the modification request, and a group of second points is generated for each element based on the group of first points and the request for modification. Then, the frame of the video stream can be transformed by modifying the elements of the object based on the group of first points and the group of second points and the grid. In such a method, the background of the modified object may also be changed or deformed by tracking and modifying the background of the modified object.
[0079] In some examples, the transformation of some areas of the object using the elements of the object can be performed by calculating the feature points of each element of the object and generating a grid based on the calculated feature points. Points are generated on the grid, and then various areas are generated based on the points. Then, the elements of the object are tracked by aligning the area of each element with the position of each element in at least one element, and the properties of the area can be modified based on the modification request, thereby transforming the frame of the video stream. Depending on the specific modification request, the properties of the mentioned area can be transformed in different ways. Such modifications may involve: changing the color of the area; removing at least some parts of the area from the frame of the video stream; including one or more new objects in the area, which is based on the modification request; and modifying or distorting the elements of the area or object. In various examples, any combination of such modifications or other similar modifications can be used. For certain models to be animated, some feature points can be selected as control points for determining the entire state space of options for model animation.
[0080] In some examples of computer animation models that use face detection to transform image data, faces are detected on images using a specific face detection algorithm (e.g., Viola-Jones). The ASM algorithm is then applied to the face region of the image to detect facial feature reference points.
[0081] Other methods and algorithms suitable for face detection can be used. For example, in some examples, landmarks are used to locate features, which represent distinguishable points that exist in most of the images considered. For example, for facial landmarks, the location of the left pupil can be used. If the initial landmark is not recognizable (for example, if the person wears an eye patch), secondary landmarks can be used. Such a landmark identification process can be used for any such object. In some examples, a set of landmarks forms a shape. The coordinates of the points in the shape can be used to represent the shape as a vector. One shape is aligned with another shape using a similarity transformation (allowing translation, scaling, and rotation) that minimizes the average Euclidean distance between the shape points. The mean shape (meanshape) is the average value of the aligned training shapes.
[0082] In some examples, the landmark search begins with a mean shape aligned with the position and size of the face determined by a global face detector. Such a search then iterates by suggesting tentative shapes by adjusting the positions of shape points through template matching of the image texture around each point, and then conforming the tentative shapes to a global shape model until convergence occurs. In some systems, individual template matches are unreliable, and the shape model pools the results of weak template matches to form a stronger overall classifier. The entire search is repeated at each level of the image pyramid from coarse to fine resolution.
[0083] The transformation system can capture an image or video stream on a client device (e.g., client device 102) and perform complex image manipulations locally on the client device 102 while maintaining an appropriate user experience, computational time, and power consumption. Complex image manipulations can include size and shape changes, emotional transformations (e.g., changing a face from a frown to a smile), state transformations (e.g., aging a subject, reducing apparent age, changing gender), style transformations, application of graphical elements, and any other suitable image or video manipulations that can be implemented by a convolutional neural network that has been configured to execute efficiently on the client device 102.
[0084] In some examples, a computer animation model for transforming image data can be used by a system in which a user can capture an image or video stream of the user (e.g., a selfie) using a client device 102 having a neural network that operates as part of a messaging client 104 operating on the client device 102. A transformation system operating within the messaging client 104 determines the presence of a face within the image or video stream and provides a modification icon associated with the computer animation model to transform the data image, or the computer animation model can be presented as associated with an interface described herein. The modification icon includes the following changes, which can be the basis for modifying the user's face in the image or video stream as part of the modification operation. Once the modification icon is selected, the transformation system initiates a process of transforming the user's image to reflect the selected modification icon (e.g., generating a smiley face on the user). Once the image or video stream is captured and the specified modification is selected, the modified image or video stream can be presented in a GUI displayed on the client device 102. The transformation system can implement a complex convolutional neural network on a portion of the image or video stream to generate and apply the selected modification. That is, a user can capture an image or video stream, and once a modification icon has been selected, the modified result can be presented in real time or near real time. Furthermore, while the video stream is being captured, the modification can be persistent and the selected modification icon remains toggled. Machine-taught neural networks can be used to implement such modifications.
[0085] The GUI presenting the modifications performed by the transformation system can supply additional interactive options for the user. Such options can be based on the interface for initiating content capture and selecting a specific computer animation model (e.g., initiated from a content creator user interface). In various examples, the modifications can be persistent after the modification icon is initially selected. The user can switch to turn the modification on or off by tapping or otherwise selecting the face being modified by the transformation system, and store it for later viewing or browsing other areas of the imaging application. In the case of multiple faces being modified by the transformation system, the user can globally turn the modification on or off by tapping or selecting a single face modified and displayed in the GUI. In some examples, each face in a group of multiple faces can be modified individually, or such modifications can be switched individually by tapping or selecting a single face or a series of individual faces displayed in the GUI.
[0086] The story table 314 stores data about a collection of messages and associated image, video or audio data that are compiled into a collection (e.g., a story or library). The creation of a particular collection can be initiated by a particular user (e.g., each user whose record is maintained in the entity table 306). A user can create a "personal story" in the form of a collection of content that has been created and sent / broadcasted by the user. To this end, the user interface of the messaging client 104 may include a user-selectable icon to enable the sending user to add specific content to his or her personal story.
[0087] A collection may also constitute a "live story" that is a collection of content from multiple users, created manually, automatically, or using a combination of manual and automatic techniques. For example, a "live story" may constitute a curated stream of user-submitted content from various locations and events. A user whose client device has location services enabled and who is at a co-located event at a particular time may be presented with the option to contribute content to a particular live story, for example, via a user interface of the messaging client 104. Live stories may be identified to a user by the messaging client 104 based on his or her location. The end result is a "live story" told from a community perspective.
[0088] Another type of content collection is called a "location story," which enables users whose client devices 102 are located in a particular geographic location (e.g., on a college or university campus) to contribute to a particular collection. In some implementations, contributions to location stories may require secondary authentication to verify that the end user belongs to a particular organization or other entity (e.g., is a student on a university campus).
[0089] As mentioned above, the video table 304 stores video data, which in one example is associated with a message whose record is stored in the message table 302. Similarly, the image table 312 stores image data associated with a message whose message data is stored in the entity table 306. The entity table 306 can associate various enhancements from the enhancement table 310 with various images and videos stored in the image table 312 and the video table 304.
[0090] The trained machine learning techniques 307 store parameters trained during training of the NLF estimation system 224. For example, the trained machine learning techniques 307 store trained parameters of one or more neural network machine learning techniques (e.g., a deep residual MLP of a neural radiance field network and / or a neural light field network). The neural network machine learning techniques can be trained to establish a relationship between a plurality of training data and corresponding ground truth data or results. For example, the neural light field network can include a machine learning model that establishes a relationship between pixel values (e.g., red, green, blue (RGB)) of one or more 2D images associated with a first view of a real-world environment and a target image associated with a second view of the real-world environment.
[0091] The training data 309 stores training data generated using a different machine learning model that implements a neural radiance field network that is different from the machine learning model used to implement the neural light field network. The training data 309 can be used by the NLF estimation system 224 to train a machine learning technique (e.g., a neural light field machine learning model) that is used to estimate pixel values of a new view of a real-world environment based on a received or captured 2D image of the real-world environment. In some cases, the training data 309 used to train the neural light field machine learning model only includes synthetic images generated using the neural radiance field network machine learning model. In some cases, the training data 309 used to train the neural light field machine learning model includes both synthetic images generated using the neural radiance field network machine learning model and captured images.
[0092] Data communication architecture
[0093] Figure 4 is a schematic diagram illustrating the structure of a message 400 according to some examples, the message 400 being generated by a messaging client 104 for transmission to another messaging client 104 or a messaging server 118. The content of a particular message 400 is used to populate a message table 302 stored in a database 126, which is accessible by a messaging server 118. Similarly, the content of the message 400 is stored in memory as "in-flight" or "in-flight" data of a client device 102 or application server 114. The message 400 is shown to include the following example components:
[0094] Message identifier 402 : a unique identifier that identifies the message 400 .
[0095] • Message text payload 404 : Text to be generated by a user via the user interface of the client device 102 and included in the message 400 .
[0096] • Message Image Payload 406: Image data captured by a camera component of the client device 102 or retrieved from a memory component of the client device 102 and included in the message 400. The image data for a message 400 sent or received may be stored in the image table 312.
[0097] Message video payload 408 : Video data captured by the camera component or retrieved from the memory component of the client device 102 and included in the message 400 . The video data for the message 400 sent or received may be stored in the video table 304 .
[0098] Message audio payload 410 : audio data captured by a microphone or retrieved from a memory component of the client device 102 and included in the message 400 .
[0099] Message enhancement data 412: Enhancement data (e.g., filters, stickers, or other annotations or enhancements) representing enhancements to be applied to the message image payload 406, message video payload 408, or message audio payload 410 of the message 400. The enhancement data 412 for a sent or received message 400 may be stored in the enhancement table 310.
[0100] Message duration parameter 414: A parameter value indicating the amount of time in seconds that the content of a message (eg, message image payload 406, message video payload 408, message audio payload 410) will be presented to or made accessible to a user via messaging client 104.
[0101] Message geolocation parameters 416: Geolocation data (e.g., latitude and longitude coordinates) associated with the content payload of the message. Multiple message geolocation parameter 416 values may be included in the payload, each of which is associated with a content item included in the content (e.g., a specific image within a message image payload 406, or a specific video within a message video payload 408).
[0102] Message story identifier 418: An identifier value that identifies one or more content collections (e.g., a "story" identified in story table 314) associated with a particular content item in message image payload 406 of message 400. For example, multiple images within message image payload 406 may each be associated with multiple content collections using an identifier value.
[0103] Message tags 420: Each message 400 may be tagged with a plurality of tags, each of which indicates the subject of the content included in the message payload. For example, where a particular image included in the message image payload 406 depicts an animal (e.g., a lion), a tag value may be included within the message tags 420 indicating the relevant animal. The tag values may be manually generated based on user input, or may be automatically generated using, for example, image recognition.
[0104] • Message sender identifier 422: An identifier (eg, a messaging system identifier, an email address, or a device identifier) that indicates the user of the client device 102 on which the message 400 was generated and from which the message 400 was sent.
[0105] • Message recipient identifier 424: An identifier (eg, a messaging system identifier, an email address, or a device identifier) that indicates the user of the client device 102 to which the message 400 is addressed.
[0106] The content (e.g., value) of various components of message 400 may be pointers to locations in a table where content data values are stored. For example, an image value in message image payload 406 may be a pointer to a location within image table 312 (or the address of a location within image table 312). Similarly, a value within message video payload 408 may point to data stored within video table 304, a value stored within message enhancement data 412 may point to data stored within enhancement table 310, a value stored within message story identifier 418 may point to data stored within story table 314, and values stored within message sender identifier 422 and message recipient identifier 424 may point to user records stored within entity table 306.
[0107] NLF Estimation System
[0108] Figure 5 is a block diagram illustrating an example NLF estimation system 224 according to some examples. The NLF estimation system 224 includes a set of components 510 that operate on a set of input data (e.g., 2D image data 501 representing a first view of a real-world environment and training data 502). The set of input data is obtained during a training phase from data stored in a database ( Figure 3 ) and obtained from the camera or storage device of the client device 102 or the messaging client 104 when using the AR / VR application. The NLF estimation system 224 includes a training data generation module 512 (which may be used only during training to generate the training data 502), an NLF prediction module 514 (e.g., a machine learning technique or a machine learning model), and a 3D model module 519 (which may implement one or more AR or VR experiences).
[0109] During training, the NLF estimation system 224 receives a given 2D training image from the training data 502. The training data 502 may include a training image and an associated camera pose associated with a plurality of training ray origins and normalized ray directions, and corresponding true pixel values. The given 2D training image may be associated with a first camera pose (a first view of a real-world environment). The first camera pose and the given 2D training image may be associated with training ray origins and normalized ray directions. The NLF estimation system 224 uses the NLF prediction module 514 to apply one or more machine learning techniques to a set of points along a training ray formed by the training ray origins and normalized ray directions to predict training pixel values for a new camera pose representing a new view of the real-world environment.
[0110] The NLF prediction module 514 retrieves the true pixel values associated with a given 2D training image from the training data 502. The NLF prediction module 514 compares the predicted training pixel values with the true pixel values. Based on the difference threshold or deviation of the comparison, the NLF prediction module 514 updates one or more coefficients or parameters of the NLF prediction module 514. The NLF prediction module 514 can determine whether the difference threshold or deviation of the comparison meets the stopping criteria. In response to determining that the difference threshold or deviation of the comparison fails to meet the stopping criteria, the NLF prediction module 514 retrieves another training image associated with a different camera pose to again estimate the pixel values of the new image representing the new view or camera pose, and updates the parameters of the NLF prediction module 514.
[0111] In response to determining that the compared difference threshold or deviation meets the stopping criteria or after a specified number of epochs or batches of training data have been processed and / or when the difference threshold or deviation reaches a specified value, the NLF prediction module 514 completes training and the parameters and coefficients of the NLF prediction module 514 are stored in the trained machine learning technique 307 (e.g., as a machine learning technique).
[0112] In some examples, the NLF prediction module 514 implements a neural light field network, e.g. Figure 6 The neural light field network 600 shown in FIG. The neural light field network 600 may include one or more deep residual MLP layers 630. Figure 7 As shown, one or more deep residual MLP layers 630 may include repeated residual MLP blocks and one or more long skips. Each residual MLP block may include a linear block and a ReLU block along with an element-wise summation block.
[0113] Return to reference Figure 6, a camera ray 610 is formed to represent a new view or new pose of a real-world environment. The camera ray 610 can be generated by accessing a given 2D image representing a specific view of the real-world environment. A new view different from the specific view of the real-world environment can be selected. The new view can include a new camera ray origin and a new camera ray direction. The camera ray origin and direction can be used to form a camera ray 610 including a plurality of points. A subset of a plurality of points (e.g., 16 points or 64 points) including a set of points 612 and 614 can be selected or sampled. The set of points 612 and 614 can be equally (uniformly) spaced along the camera ray 610, or can be randomly selected from a plurality of points located along the camera ray 610. In some examples, a subset of a plurality of points including the set of points 612 and 614 can be selected or sampled based on stratified sampling during training of the machine learning model.
[0114] In some examples, the neural light field network 600 obtains the values of the set of points 612 and 614 of the camera ray 610. The neural light field network 600 can then combine, sum, or concatenate the set of points 612 and 614 to generate input data. The input data is processed by one or more deep residual MLP layers 630 in a single forward pass to generate predicted pixel values 640 associated with the camera ray 610. After predicting the plurality of pixel values 640, a new predicted image representing a new view or new pose of the real-world environment can be formed from the predicted pixel values 640.
[0115] In this way, the NLF prediction module 514 can generate a scene represented as a light field parameterized by a neural network rather than a radiance field. The neural network FΘ learns a mapping function (e.g., Fθ) directly from 5D directional rays to their target 3D RGB values. Θ :R 5 ->R 3 ). Using NLF technology instead of NRF (neural radiance field network) to estimate RGB values has the advantage of being more straightforward to implement. This is because the output of the NRF network is the radiance values of multiple sampling points, which need to be processed by light matching technology to estimate the actual pixel values from the radiance values. When the NLF technology is applied to the sampling points, no additional processing is required because the NLF technology directly provides the corresponding RGB pixel values. In addition, the NRF network (as shown below) generates radiance values by repeatedly applying the NRF network to each corresponding point sampled from the camera light. This repeated application consumes a lot of resources and is inefficient compared to performing a single pass on the result of connecting multiple points of the light. Therefore, the NLF technology can provide estimated pixel values at a rate several orders of magnitude faster than the application of the NRF network to the same camera light.
[0116] The pixel values predicted by the NLF prediction module 514 for a particular view of the scene (e.g., a new view of the real-world environment) are provided to the 3D model module 519. In some cases, the predicted pixel values are processed together to generate a new image representing the new view of the real-world environment. The new image and one or more previously captured 2D images of the real-world environment are provided to the 3D model module 519. The 3D model module 519 performs 3D reconstruction on the received images to generate a 3D model representing the real-world environment. The 3D model can be used to add one or more virtual elements to the real-world environment depicted in the new image and / or captured image, thereby providing an AR or VR experience for the end user on the client device 102.
[0117] In some cases, the NLF prediction module 514 receives multiple images captured by the camera of the client device 102. The NLF prediction module 514 can determine the camera pose or view associated with the multiple images by specifying different light origins and directions for each image. The light origin and direction can be received based on metadata associated with the captured image. The NLF prediction module 514 can receive input from a user for selecting a new view of the real-world environment that is not included in the multiple images. That is, the new view can be different from the light origin and direction of the captured image. In response, the NLF prediction module 514 can generate new light corresponding to the new view based on one or more captured images. Then, the NLF prediction module 514 can predict one or more pixel values of the new image corresponding to the new view to generate the new image. The NLF prediction module 514 can provide the new image and the captured image to the 3D model module 519 to generate a 3D model of the real-world environment.
[0118] In some cases, the training data 502 used to train the NLF prediction module 514 is generated by the training data generation module 512. The training data generation module 512 may be implemented as follows: Figure 8 NRF network 800 shown. Specifically, training data generation module 512 can be applied to a set of 2D images to generate training data. NRF network 800 (e.g., neural radiance field network) can include a trained neural network that is configured to generate an output representing the radiation of one or more sampling points corresponding to a specific light. The pixel value of a specific light can be generated by alpha compositing a plurality of points including the sampling points along the specific light corresponding to the corresponding radiation value.
[0119] In some examples, the training data 502 is generated by receiving a first 2D image in a set of 2D images. Based on the first 2D image, a ray origin and a normalized direction of a camera pose associated with a second 2D image are selected, and the second 2D image is associated with a camera pose different from the first 2D image. The ray origin and the normalized direction of the camera pose can be randomly selected based on a uniform distribution. A second machine learning model (e.g., an NRF network 800) is applied to the ray origin and the normalized direction of the camera pose associated with the second 2D image to predict the pixel value of the second 2D image. Specifically, the NRF network 800 is applied to each of a plurality of points 812 and 814 of a ray formed along the ray origin and the normalized direction of the camera pose associated with the second 2D image to generate a plurality of radiance values. In some cases, a shallow MLP network is applied to the first point 812 to generate a first radiance value, and then a shallow MLP network is applied to the second point 814 to generate a second radiance value. The plurality of radiance values 820 are collectively processed, for example, by an alpha synthesis network 830 to generate RGB values for pixels associated with a ray formed by the ray origin and normalized direction for the camera pose associated with the second 2D image.
[0120] In some examples, the ray origin is selected (e.g., x o ,y o ,z o ) and normalized directions (e.g. x d ,y d ,z d ) so that they obey a uniform distribution within the bounding box for the following 6D input:
[0121]
[0122] Given a trained NRF network 800 (F Θ* ), the target RGB value 840 can be known by querying the NRF network 800 (e.g., where Θ* represents the converged model parameters. A slice of the training data can be represented by a vector of nine numbers Then, the neural light field network 600 can be trained based on the loss function corresponding to the following formula:
[0123]
[0124] In some examples, the neural light field network 600 is trained on light that is more difficult to regress during its learning (e.g., corresponding to higher frequency details). To this end, a pool of hard examples is provided. Hard examples can be defined based on the larger loss in the above loss function. In each iteration, the loss of each sample in the batch is sorted in ascending order. From the sorted losses, the highest r (or a predefined percentage constant or number) are added to the hard example pool. In each iteration, the same amount or number of r hard samples can be randomly selected from the pool to enhance the training batch. This can improve the convergence of the network.
[0125] Fig. 9 900 is a flow chart of a process 900 performed by the NLF estimation system 224 according to some examples. Although the flow chart may describe the operations as sequential processes, many operations may be performed in parallel or concurrently. In addition, the order of the operations may be rearranged. A process terminates when its operations are completed. A process may correspond to a method, a procedure, etc. The steps of a method may be performed in whole or in part, may be performed in combination with some or all steps in other methods, and may be performed by any number of different systems or any part thereof (e.g., a processor included in any system).
[0126] At operation 901 , the NLF estimation system 224 (eg, the client device 102 or a server) receives a set of two-dimensional (2D) images representing a first view of a real-world environment (as described above).
[0127] At operation 902, the NLF estimation system 224 applies a machine learning model including a neural light field network to the set of 2D images to predict pixel values of a target image for representing a second view of the real-world environment, the machine learning model being trained to map light ray origins and directions directly to given pixel values (as described above).
[0128] At operation 903 , the NLF estimation system 224 generates a three-dimensional (3D) model of the real-world environment based on the set of 2D images and the predicted target image (as described above).
[0129] Machine Architecture
[0130] Fig.10It is a diagrammatic representation of a machine 1000, within which instructions 1008 (e.g., software, programs, applications, applet, APP or other executable code) for causing the machine 1000 to perform any one or more of the methods discussed herein can be executed. For example, the instructions 1008 can cause the machine 1000 to perform any one or more of the methods described herein. The instructions 1008 convert a general unprogrammed machine 1000 into a specific machine 1000 that is programmed to perform the functions described and shown in the described manner. The machine 1000 can operate as a standalone device, or can be coupled (e.g., networked) to other machines. In a networked deployment, the machine 1000 can operate with the performance of a server machine or a client machine in a server-client network environment, or operate as a peer machine in a peer-to-peer (or distributed) network environment. The machine 1000 may include, but is not limited to, a server computer, a client computer, a PC, a tablet computer, a laptop computer, a netbook, an STB, a PDA, an entertainment media system, a cellular phone, a smart phone, a mobile device, a wearable device (e.g., a smart watch), a smart home device (e.g., a smart appliance), other smart devices, a web appliance, a network router, a network switch, a network bridge, or any machine capable of sequentially or otherwise executing instructions 1008 specifying actions to be taken by the machine 1000. In addition, although only a single machine 1000 is shown, the term "machine" should also be considered to include a collection of machines that individually or jointly execute instructions 1008 to perform any one or more of the methods discussed herein. For example, the machine 1000 may include any of the client devices 102 or a number of server devices that form part of the messaging server system 108. In some examples, the machine 1000 may also include both a client system and a server system, wherein certain operations of a particular method or algorithm are performed on the server side and certain operations of the particular method or algorithm are performed on the client side.
[0131] The machine 1000 may include a processor 1002, a memory 1004, and an input / output (I / O) component 1038 that may be configured to communicate with each other via a bus 1040. In an example, the processor 1002 (e.g., a CPU, a reduced instruction set computing (RISC) processor, a complex instruction set computing (CISC) processor, a 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, a processor 1006 that executes instructions 1008 and a processor 1010. The term "processor" is intended to include multi-core processors, which may include two or more independent processors (sometimes referred to as "cores") that may execute instructions simultaneously. Although Fig.10Multiple processors 1002 are shown, but the machine 1000 may include a single processor with a single core, a single processor with multiple cores (e.g., a multi-core processor), multiple processors with a single core, multiple processors with multiple cores, or any combination thereof.
[0132] The memory 1004 includes a main memory 1012, a static memory 1014, and a storage unit 1016, which are all accessible by the processor 1002 via the bus 1040. The main memory 1004, the static memory 1014, and the storage unit 1016 store instructions 1008 that embody any one or more of the methodologies or functions described herein. The instructions 1008 may also reside, completely or partially, within the main memory 1012, within the static memory 1014, within the machine-readable medium 1018 within the storage unit 1016, within at least one of the processors 1002 (e.g., within a cache memory of a processor), or within any suitable combination thereof during execution thereof by the machine 1000.
[0133] I / O components 1038 may include various components that receive input, provide output, generate output, send information, exchange information, capture measurements, etc. The specific I / O components 1038 included in a particular machine will depend on the type of machine. For example, a portable machine such as a mobile phone may include a touch input device or other such input mechanism, while a headless server machine may not include such a touch input device. It should be appreciated that I / O components 1038 may include Fig.10 1024 and 1026. In various examples, the I / O components 1038 may include a user output component 1024 and a user input component 1026. The user output component 1024 may include a visual component (e.g., a display such as a plasma display panel (PDP), a light emitting diode (LED) display, a liquid crystal display (LCD), a projector, or a cathode ray tube (CRT)), an acoustic component (e.g., a speaker), a tactile component (e.g., a vibration motor, a resistance mechanism), other signal generators, etc. The user input component 1026 may include an alphanumeric input component (e.g., a keyboard, a touch screen configured to receive alphanumeric input, an optical keyboard, or other alphanumeric input component), a point-based input component (e.g., a mouse, a touch pad, a trackball, a joystick, a motion sensor, or other pointing instrument), a tactile input component (e.g., a physical button, a touch screen or other tactile input component that provides the location and force of a touch or touch gesture), an audio input component (e.g., a microphone), etc.
[0134] In other examples, I / O component 1038 may include biometric component 1028, motion component 1030, environment component 1032, or position component 1034, as well as various other components. For example, biometric component 1028 includes components for detecting expressions (e.g., hand expressions, facial expressions, voice expressions, body postures, or eye tracking), measuring biosignals (e.g., blood pressure, heart rate, body temperature, sweat, or brain waves), identifying people (e.g., voice recognition, retinal recognition, facial recognition, fingerprint recognition, or EEG-based recognition), etc. Motion component 1030 includes acceleration sensor components (e.g., accelerometers), gravity sensor components, and rotation sensor components (e.g., gyroscopes).
[0135] Environmental components 1032 include, for example, one or more cameras (with still image / photo and video capabilities), lighting sensor components (e.g., photometers), temperature sensor components (e.g., one or more thermometers that detect ambient temperature), humidity sensor components, pressure sensor components (e.g., barometers), acoustic sensor components (e.g., one or more microphones that detect background noise), proximity sensor components (e.g., infrared sensors that detect nearby objects), gas sensors (e.g., gas detection sensors that detect concentrations of hazardous gases for safety or measure pollutants in the atmosphere), or other components that can provide indications, measurements, or signals corresponding to the surrounding physical environment.
[0136] With respect to cameras, client device 102 can have a camera system including, for example, a front-facing camera on a front surface of client device 102 and a rear-facing camera on a rear surface of client device 102. The front-facing camera can, for example, be used to capture still images and videos (e.g., “selfies”) of a user of client device 102, which can then be enhanced using the enhancement data (e.g., filters) described above. The rear-facing camera can, for example, be used to capture still images and videos in a more traditional camera mode, which can similarly be enhanced using the enhancement data. In addition to the front-facing camera and the rear-facing camera, client device 102 can also include a 360° camera for capturing 360° photos and videos.
[0137] Additionally, the camera system of the client device 102 may include dual rear cameras (e.g., a main camera and a depth sensing camera), or even triple, quad, or quintuple rear camera configurations on the front and back sides of the client device 102. For example, these multi-camera systems may include a wide-angle camera, an ultra-wide-angle camera, a telephoto camera, a macro camera, and a depth sensor.
[0138] The location component 1034 may include a positioning sensor component (e.g., a GPS receiver component), an altitude sensor component (e.g., an altimeter or a barometer that detects air pressure that can obtain altitude), an orientation sensor component (e.g., a magnetometer), and the like.
[0139] A variety of technologies may be used to achieve communication. The I / O components 1038 also include a communication component 1036 that is operable to couple the machine 1000 to the network 1020 or the device 1022 via a corresponding coupling or connection. For example, the communication component 1036 may include a network interface component or other suitable device that interfaces with the network 1020. In other examples, the communication component 1036 may include a wired communication component, a wireless communication component, a cellular communication component, a near field communication (NFC) component, a Bluetooth component (e.g., Bluetooth low energy), a Wi-Fi component, and other communication components that provide communication via other modalities. The device 1022 may be any peripheral device of another machine or a variety of peripheral devices (e.g., a peripheral device coupled via USB).
[0140] In addition, the communication component 1036 can detect an identifier or include a component that can be operated to detect an identifier. For example, the communication component 1036 can include a radio frequency identification (RFID) tag reader component, an NFC smart tag detection component, an optical reader component (e.g., an optical sensor for detecting a one-dimensional barcode such as a universal product code (UPC) barcode, a multi-dimensional barcode such as a quick response (QR) code, an Aztec code, a data matrix, a data symbol (Dataglyph), a maximum code (MaxiCode), PDF417, Ultra Code, UCC RSS-2D barcode, and other optical codes) or an acoustic detection component (e.g., a microphone for identifying an audio signal of a tag). In addition, various information can be derived via the communication component 1036, such as a location derived via Internet Protocol (IP) geolocation, a location derived via Wi-Fi signal triangulation, a location derived via detection of an NFC beacon signal that can indicate a specific location, etc.
[0141] Various memories (e.g., main memory 1012, static memory 1014, and memory of processor 1002) and storage unit 1016 may store one or more sets of instructions and data structures (e.g., software) implemented or used by any one or more of the methods or functions described herein. When executed by processor 1002, these instructions (e.g., instructions 1008) cause various operations to implement the disclosed examples.
[0142] The instructions 1008 may be sent or received over the network 1020 via a network interface device (e.g., a network interface component included in the communications component 1036) using a transmission medium and using any of several well-known transmission protocols (e.g., HTTP). Similarly, the instructions 1008 may be sent or received via a coupling (e.g., a peer-to-peer coupling) with the device 1022 using a transmission medium.
[0143] Software Architecture
[0144] Fig.11 1100 is a block diagram illustrating a software architecture 1104 that may be installed on any one or more of the devices described herein. The software architecture 1104 is supported by hardware, such as a machine 1102 including a processor 1120, a memory 1126, and an I / O component 1138. In this example, the software architecture 1104 may be conceptualized as a stack of layers, where each layer provides specific functionality. The software architecture 1104 includes layers such as an operating system 1112, a library 1110, a framework 1108, and an application 1106. In operation, the application 1106 invokes an API call 1150 through the software stack and receives a message 1152 in response to the API call 1150.
[0145] The operating system 1112 manages hardware resources and provides public services. The operating system 1112 includes, for example, a core 1114, a service 1116, and a driver 1122. The core 1114 serves as an abstraction layer between the hardware and other software layers. For example, the core 1114 provides functions such as memory management, processor management (e.g., scheduling), component management, networking, and security settings. The service 1116 can provide other public services for other software layers. The driver 1122 is responsible for controlling the underlying hardware or interfacing with the underlying hardware. For example, the driver 1122 may include a display driver, a camera driver, a Bluetooth or Bluetooth low energy driver, a flash drive, a serial communication driver (e.g., a USB driver), a WI-FI driver, an audio driver, a power management driver, etc.
[0146] The libraries 1110 provide a common low-level infrastructure used by the applications 1106. The libraries 1110 may include a system library 1118 (e.g., a C standard library) that provides functions such as memory allocation functions, string manipulation functions, mathematical functions, etc. In addition, the libraries 1110 may include an API library 1124, such as a media library (e.g., a library for supporting the presentation and manipulation of various media formats, such as Moving Picture Experts Group-4 (MPEG4), Advanced Video Coding (H.264 or AVC), Moving Picture Experts Group Layer-3 (MP3), Advanced Audio Coding (AAC), Adaptive Multi-Rate (AMR) audio codec, Joint Photographic Experts Group (JPEG or JPG), or Portable Network Graphics (PNG)), a graphics library (e.g., an OpenGL framework for rendering in 2D and 3D in graphics content on a display), a database library (e.g., SQLite providing various relational database functions), a web library (e.g., WebKit providing web browsing functions), etc. The library 1110 may also include various other libraries 1128 to provide many other APIs to the application 1106 .
[0147] The framework 1108 provides a common high-level infrastructure used by the applications 1106. For example, the framework 1108 provides various GUI functions, high-level resource management, and high-level positioning services. The framework 1108 can provide a wide range of other APIs that can be used by the applications 1106, some of which can be specific to a particular operating system or platform.
[0148] In the example, applications 1106 may include home applications 1136, contact applications 1130, browser applications 1132, book reader applications 1134, positioning applications 1142, media applications 1144, message applications 1146, game applications 1148, and various other applications such as external applications 1140. Applications 1106 are programs that execute functions defined in programs. Various programming languages can be used to create one or more of the applications 1106 constructed in various ways, 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 specific example, external applications 1140 (e.g., applications developed using ANDROID or IOS SDK by entities other than the supplier of a specific platform) can be mobile software running on mobile operating systems such as IOS, ANDROID, WINDOWS Phone, or other mobile operating systems. In this example, external applications 1140 can activate API calls 1150 provided by the operating system 1112 to facilitate the functions described herein.
[0149] Glossary
[0150] "Carrier signal" refers to any intangible medium that can store, encode or carry instructions for execution by a machine and includes digital or analog communications signals or other intangible media that facilitates communication of such instructions. Instructions may be sent or received over a network using a transmission medium via a network interface device.
[0151] "Client Device" refers to any machine that interfaces with a communications network to obtain resources from one or more server systems or other client devices. A client device may be, but is not limited to, a mobile phone, desktop computer, laptop computer, PDA, smart phone, tablet computer, ultrabook, netbook, laptop computer, multiprocessor system, microprocessor-based or programmable consumer electronics, game console, STB, or any other communications device that a user may use to access a network.
[0152] "Communications network" refers to one or more parts of a network, which may be an ad hoc network, an intranet, an extranet, a virtual private network (VPN), a local area network (LAN), a wireless LAN (WLAN), a wide area network (WAN), a wireless WAN (WWAN), a metropolitan area network (MAN), the Internet, a part of the Internet, a part of the public switched telephone network (PSTN), a plain old telephone service (POTS) network, a cellular telephone network, a wireless network, A network, other type of network, or a combination of two or more such networks. For example, the network or a portion of the 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 type of cellular or wireless coupling. In this example, the coupling may implement any of various types of data transmission technologies, such as single carrier radio transmission technology (1xRTT), evolution data optimized (EVDO) technology, general packet radio service (GPRS) technology, enhanced data rates for GSM evolution (EDGE) technology, the third generation partnership project (3GPP) including 3G, fourth generation wireless (4G) networks, universal mobile telecommunications system (UMTS), high speed packet access (HSPA), world wide interoperability for microwave access (WiMAX), long term evolution (LTE) standards, other data transmission technologies defined by various standard setting organizations, other long distance protocols, or other data transmission technologies.
[0153] "Component" means a logic, device, or physical entity having boundaries defined by function or subroutine calls, branch points, APIs, or other techniques that provide partitioning or modularization of specific processing or control functions. Components may be combined with other components via their interfaces to perform machine processing. A component may be a packaged functional hardware unit designed for use with other components and may be part of a program that typically performs a specific one of the related functions.
[0154] Components may constitute software components (e.g., code implemented on a machine-readable medium) or hardware components. A "hardware component" is a tangible unit that is capable of performing certain operations and that can be configured or arranged in some physical manner. In various example implementations, one or more computer systems (e.g., stand-alone computer systems, client computer systems, or server computer systems) or one or more hardware components of a computer system (e.g., a processor or group of processors) may be configured by software (e.g., an application or application portion) as hardware components that operate to perform certain operations described herein.
[0155] The hardware components may also be implemented mechanically, electronically, or in any suitable combination thereof. For example, the hardware components may include a dedicated circuit or logic that is permanently configured to perform certain operations. The hardware components may be a dedicated processor, such as a field programmable gate array (FPGA) or an ASIC. The hardware components may also include a programmable logic or circuit system that is temporarily configured to perform certain operations by software. For example, the hardware components may include software executed by a general-purpose processor or other programmable processor. Once configured by such software, the hardware components become a specific machine (or a specific component of a machine) that is uniquely customized to perform the configured function, and are no longer a general-purpose processor. It should be recognized that it can be determined for cost and time considerations whether to implement the hardware components mechanically in a dedicated and permanently configured circuit system or in a temporarily configured (e.g., configured by software) circuit system. Therefore, the phrase "hardware components" (or "hardware-implemented components") should be understood to include tangible entities, i.e., entities that are physically constructed, permanently configured (e.g., hardwired) or temporarily configured (e.g., programmed) to operate in some way or perform certain operations described herein.
[0156] Consider an example where hardware components are temporarily configured (e.g., programmed), without configuring or instantiating each of the hardware components at any one time. For example, where the hardware components include a general-purpose processor that is configured by software to be a special-purpose processor, the general-purpose processor can be configured to be a different special-purpose processor (e.g., including different hardware components) at different times. The software accordingly configures one or more specific processors to, for example, constitute a specific hardware component at one time and to constitute a different hardware component at a different time.
[0157] Hardware components can provide information to other hardware components and receive information from other hardware components. Therefore, the described hardware components can be considered to be communicatively coupled. In the case of multiple hardware components simultaneously, communication can be realized by signal transmission between two or more hardware components (for example, by appropriate circuits and buses). In the example where multiple hardware components are configured or instantiated at different times, the communication between such hardware components can be realized, for example, by storing information in a memory structure accessible to multiple hardware components and retrieving information in the memory structure. For example, a hardware component can perform an operation and store the output of the operation in a memory device to which it is communicatively coupled. Then, another hardware component can access the memory device at a subsequent time to retrieve the stored output and process it. The hardware component can also initiate communication with an input device or an output device, and can operate on resources (for example, a collection of information).
[0158] The various operations of the example methods described herein may be performed at least in part by one or more processors that are temporarily configured (e.g., by software) or permanently configured to perform related operations. Whether temporarily configured or permanently configured, such a processor may constitute a processor-implemented component that operates to perform one or more operations or functions described herein. As used herein, a "processor-implemented component" refers to a hardware component implemented using one or more processors. Similarly, the method described herein may be implemented at least in part by a processor, wherein a specific one or more processors are examples of hardware. For example, at least some operations of the method may be performed by one or more processors 1002 or a processor-implemented component. In addition, the one or more processors may 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 in the operation may be performed by a group of computers (as an example of a machine including a processor), wherein these operations may be accessed via a network (e.g., the Internet) and via one or more appropriate interfaces (e.g., API). The execution of certain operations may be distributed between processors, not just residing in a single machine, but deployed across several machines. In some examples, the processor or processor-implemented components may be located in a single geographic location (e.g., in a home environment, an office environment, or a server farm). In other examples, the processor or processor-implemented components may be distributed across several geographic locations.
[0159] "Computer-readable storage media" refers to both 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" mean the same thing and may be used interchangeably in this disclosure.
[0160] "Ephemeral messages" are messages that are accessible for a limited duration. Ephemeral messages can be text, images, videos, etc. The access time for ephemeral messages can be set by the sender of the message. Alternatively, the access time can be a default setting or a setting specified by the recipient. Regardless of the setting technique, the message is transient.
[0161] "Machine storage media" refers to a single or multiple storage devices and media (e.g., centralized or distributed databases, and associated caches and servers) that store executable instructions, routines, and data. Thus, the 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), FPGA, and flash memory devices; magnetic disks, such as internal hard disks and removable disks; magneto-optical disks; and CD-ROM and DVD-ROM disks. The terms "machine storage media," "device storage media," and "computer storage media" mean the same thing and may be used interchangeably in this disclosure. The terms "machine storage media," "computer storage media," and "device storage media" expressly exclude carrier waves, modulated data signals, and other such media (at least some of which are encompassed by the term "signal media").
[0162] “Non-transitory computer-readable storage medium” refers to tangible media capable of storing, encoding, or carrying instructions to be executed by a machine.
[0163] "Signal medium" refers to any intangible medium that can store, encode or carry instructions executed 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 deemed to include any form of modulated data signals, carrier waves, etc. The term "modulated data signal" refers to a signal that has one or more of its various characteristics set or changed in such a manner as to encode information in the signal. The terms "transmission medium" and "signal medium" refer to the same thing and may be used interchangeably in this disclosure.
[0164] Changes and modifications may be made to the disclosed examples without departing from the scope of the present disclosure. These and other changes or modifications are intended to be included within the scope of the present disclosure as expressed in the appended claims.
Claims
1. A method comprising: receiving, by one or more processors, a set of two-dimensional (2D) images representing a first view of a real-world environment; applying, by the one or more processors, a machine learning model comprising a neural light field network to the set of 2D images to predict pixel values of a target image representing a second view of the real-world environment, the machine learning model being trained to directly map a light ray origin and direction to a given pixel value; as well as A three-dimensional (3D) model of the real world environment is generated by the one or more processors based on the set of 2D images and the predicted target image.
2. The method according to claim 1, wherein: The machine learning model includes a deep residual multi-layer perceptron (MLP) network.
3. The method according to any one of claims 1 to 2, further comprising: selecting a ray origin and direction associated with a second view of the real world environment; generating, using a given 2D image in the set of 2D images representing a first view of the real-world environment, a ray based on the ray origin and direction corresponding to a second view of the real-world environment; and A plurality of points along the ray are sampled.
4. The method according to claim 3, wherein: The plurality of points are evenly spaced apart along the ray.
5. The method according to any one of claims 3 to 4, wherein: During training of the machine learning model, the plurality of points are randomly sampled based on stratified sampling.
6. The method according to any one of claims 3 to 5, further comprising: connecting the plurality of points to generate input data; as well as The input data is processed using the machine learning model to predict one of the pixel values of the 2D object image.
7. The method according to any one of claims 1 to 6, further comprising training the machine learning model by performing operations comprising: Receiving training data, the training data comprising training images and associated camera poses and corresponding true pixel values, the training images and associated camera poses being associated with a plurality of training ray origins and normalized ray directions; Obtaining a training light origin and a normalized light direction of a first training image and an associated camera device posture in the training images; applying the machine learning model to a set of points along a training ray formed by the training ray origin and the normalized ray direction to predict training pixel values; retrieving true pixel values associated with the first training image; Calculating the deviation between the predicted training pixel value and the true pixel value; and One or more parameters of the machine learning model are updated based on the bias.
8. The method according to claim 7, wherein: The machine learning model includes a first machine learning model, and the method further includes: applying a second machine learning model to a set of 2D images to generate the training data.
9. The method according to any one of claims 1 to 8, wherein: The second machine learning model includes a trained neural radiance field network, which is configured to generate an output representing the radiance of a sampling point corresponding to a specific light ray, wherein the pixel value of the specific light ray is generated by alpha compositing a plurality of points along the specific light ray including the sampling point corresponding to corresponding radiance values.
10. The method according to any one of claims 1 to 9, further comprising: receiving a first 2D image from the set of 2D images; Based on the first 2D image, selecting a ray origin and a normalized direction of a camera pose associated with a second 2D image, the second 2D image being associated with a different camera pose than the first 2D image; Applying the second machine learning model to the ray origin and normalized direction of the camera pose associated with the second 2D image to predict pixel values of the second 2D image, wherein applying the second machine learning model comprises: applying the second machine learning model to each of a plurality of points along a ray formed by the ray origin and the normalized direction of the camera pose associated with the second 2D image to generate a plurality of radiance values; and Alpha synthesis is performed on the plurality of radiance values to predict pixel values of the second 2D image.
11. The method according to claim 10, wherein: The light origin and normalized direction of the camera device posture are randomly selected based on a uniform distribution.
12. The method according to any one of claims 10 to 11, wherein: The training data does not include the set of 2D images.
13. The method according to any one of claims 1 to 12, further comprising: generating a plurality of losses for the training data; sorting the plurality of losses in ascending order; as well as A number of samples in the ordered plurality of losses that exceeds a specified threshold is identified.
14. The method according to claim 13, further comprising: The training data is augmented with additional training data corresponding to the number of samples.
15. The method according to any one of claims 1 to 14, further comprising: Based on the 3D model of the real-world environment, virtual elements associated with an augmented reality or virtual reality experience are displayed on a client device within a video including the set of 2D images.
16. A system comprising The processor of the client device; as well as A memory component, wherein the memory component stores instructions, and when the instructions are executed by the processor, the processor performs operations, the operations comprising: receiving a set of two-dimensional (2D) images representing a first view of a real-world environment; applying a machine learning model comprising a neural light field network to the set of 2D images to predict pixel values of a target image representing a second view of the real-world environment, the machine learning model being trained to directly map a light ray origin and direction to a given pixel value; and A three-dimensional (3D) model of the real-world environment is generated based on the set of 2D images and the predicted target image.
17. The system of claim 16, the operations further comprising: selecting a ray origin and direction associated with a second view of the real world environment; generating, using a given 2D image in the set of 2D images representing a first view of the real-world environment, a ray based on the ray origin and direction corresponding to a second view of the real-world environment; and A plurality of points along the ray are sampled.
18. The system according to any one of claims 16-17, wherein: The plurality of points are evenly spaced apart along the ray.
19. The system according to any one of claims 16 to 18, wherein: During training of the machine learning model, the plurality of points are randomly sampled based on stratified sampling.
20. A non-transitory computer-readable storage medium having instructions stored thereon, wherein when the instructions are executed by a processor of a client device, the processor performs operations comprising: receiving a set of two-dimensional (2D) images representing a first view of a real-world environment; applying a machine learning model comprising a neural light field network to the set of 2D images to predict pixel values of a target image representing a second view of the real-world environment, the machine learning model being trained to directly map a light ray origin and direction to a given pixel value; and A three-dimensional (3D) model of the real-world environment is generated based on the set of 2D images and the predicted target image.
Citation Information
Cited By
3D modeling based on neural light field
US12450822B2
