Facebook Patent | Multi-Device Mapping And Collaboration In Augmented-Reality Environments

Patent: Multi-Device Mapping And Collaboration In Augmented-Reality Environments

Publication Number: 20200066045

Publication Date: 20200227

Applicants: Facebook

Abstract

In one embodiment, a method includes receiving first information from a sensor associated with a first computing device, wherein the first information comprises information associated with first images captured at the first sensor; receiving second information from a second computing device, wherein the second information comprises information associated with second images captured at a sensor associated with the second computing device; identifying first points within the first images; identifying second points within the second images; and relocalizing the first and second computing devices within a shared augmented-reality environment by defining coordinate spaces based on the images and combining the coordinate spaces based on identified shared points.

TECHNICAL FIELD

[0001] This disclosure generally relates to augmented or virtual reality environments.

BACKGROUND

[0002] Augmented Reality (AR) effects are computer-generated visual effects (e.g., images and animation) that are superimposed or integrated into a user’s view of a real-world scene. Certain AR effects may be configured to track objects in the real world. For example, a computer-generated unicorn may be placed on a real-world table as captured in a video. As the table moves in the captured video (e.g., due to the camera moving or the table being carried away), the generated unicorn may follow the table so that it continues to appear on top of the table. To achieve this effect, an AR application may use tracking algorithms to track the positions and/or orientations of objects appearing in the real-world scene and use the resulting tracking data to generate the appropriate AR effect. Since AR effects may augment the real-world scene in real-time or near real-time while the scene is being observed, tracking data may need to be generated in real-time or near real-time so that the AR effect appears as desired.

[0003] A social-networking system, which may include a social-networking website, may enable its users (such as persons or organizations) to interact with it and with each other through it. The social-networking system may, with input from a user, create and store in the social-networking system a user profile associated with the user. The user profile may include demographic information, communication-channel information, and information on personal interests of the user. The social-networking system may also, with input from a user, create and store a record of relationships of the user with other users of the social-networking system, as well as provide services (e.g., wall posts, photo-sharing, event organization, messaging, games, or advertisements) to facilitate social interaction between or among users.

SUMMARY OF PARTICULAR EMBODIMENTS

[0004] In particular embodiments, a processor of a first computing device may receive first information from one or more sensors associated with the first computing device. This may include information about images captured at the sensors. For example, the sensors may include a camera that captures photo images. In particular embodiments, a computing system (e.g., the first computing device, a server) may receive second information from a second computing device. The second information may include information about images captured at one or more sensors associated with the second computing device. In particular embodiments, a computing system may identify one or more first points within the first images and one or more second points within the second images. These points may correspond to features of interest in the images. In particular embodiments, a computing system may “relocalize” the first computing device and the second computing device within a shared AR environment. This relocalization process may be an initialization or re-initialization process that may be used to orient an AR device within an AR environment.

[0005] The disclosed invention may provide several technological benefits. For example, the relocalization of multiple devices may allow for collaboration among multiple devices in mapping a real-world environment onto an AR environment and tracking objects within the real-world or AR environments. This collaboration among multiple devices may markedly improve the mapping process by reducing the time and effort it takes to map an environment. This improvement may be particularly pronounced and its benefits may compound significantly in cases where the real-world environment that is to be mapped is very large. As an example and not by way of limitation, the time and effort involved with mapping of a large football stadium may be significantly reduced if thirty users can relocalize and map the stadium separately to create a shared AR environment that includes all the points that they each mapped out separately.

[0006] In particular embodiments, a user may share or present an AR content item to other users within an AR environment. In particular embodiments, a first computing device associated with a first user may render an AR environment on a display associated with the first computing device. In particular embodiments, a computing system may determine that the first computing device is authorized to access a particular AR content item associated with a location within the AR environment based on social-networking information associated with the first user. The particular AR content item may be associated with a second user. In particular embodiments, the first computing device may receive information configured to render the AR content item at the associated location within the AR environment. In particular embodiments, the AR content item may be rendered on the display associated with the first computing device, such that it appears within the AR environment.

[0007] In particular embodiments, the AR content item may be integrated into the AR environment and tied to particular locations or objects identified in the environment. By so doing, a more immersive experience may be created for a user viewing an AR environment, such that the AR content may be perceived as being an integral part of the real-world environment being modeled by the AR environment. In particular embodiments, he first user may be able to view or interact with the AR content item in intuitive ways (e.g., picking them up, moving them, modifying them, etc.).

[0008] In particular embodiments, an AR environment may be rendered. In particular embodiments, one or more objects (e.g., a face, an animal, a real-world inanimate object, a virtual object, etc.) may be identified within an AR environment rendered on a display associated with a first computing device associated with a first user. In particular embodiments, a computing system may determine one or more suggested AR content items for the first user. In particular embodiments, the determination as to what content to suggest may be based on social-networking information associated with the first user. In particular embodiments, the determination as to what content to suggest may be based on characteristics associated with the objects identified in the AR environment. In particular embodiments, the suggested AR content item may be presented as suggestions on a display associated with the first computing device. A user may then be able to select a suggested AR content item and interact with it (e.g., by placing it at a location within the AR environment).

[0009] By providing a suggestion tool that intelligently determines optimal AR content that is tailored to the user and the circumstances, users may be encouraged to place AR content within an AR environment or otherwise interact with objects in the AR environment. The suggestion tool may allow for discovery of content that may be particularly relevant to a user (e.g., content created by friends of the user, content tailored to the user’s interests, etc.) at any given time.

[0010] The embodiments disclosed herein are only examples, and the scope of this disclosure is not limited to them. Particular embodiments may include all, some, or none of the components, elements, features, functions, operations, or steps of the embodiments disclosed herein. Embodiments according to the invention are in particular disclosed in the attached claims directed to a method, a storage medium, a system and a computer program product, wherein any feature mentioned in one claim category, e.g. method, can be claimed in another claim category, e.g. system, as well. The dependencies or references back in the attached claims are chosen for formal reasons only. However any subject matter resulting from a deliberate reference back to any previous claims (in particular multiple dependencies) can be claimed as well, so that any combination of claims and the features thereof are disclosed and can be claimed regardless of the dependencies chosen in the attached claims. The subject-matter which can be claimed comprises not only the combinations of features as set out in the attached claims but also any other combination of features in the claims, wherein each feature mentioned in the claims can be combined with any other feature or combination of other features in the claims. Furthermore, any of the embodiments and features described or depicted herein can be claimed in a separate claim and/or in any combination with any embodiment or feature described or depicted herein or with any of the features of the attached claims.

BRIEF DESCRIPTION OF THE DRAWINGS

[0011] FIG. 1 illustrates example embodiments of multiple devices configured to communicate information with each other.

[0012] FIG. 2 illustrates an example real-world scene captured within an image or video frame.

[0013] FIG. 3 illustrates an example of an image with identified points associated with certain objects of interest.

[0014] FIG. 4 illustrates another example of an image with identified points associated with an object of interest.

[0015] FIG. 5 illustrates an example abstraction of the concept of relocalizing multiple coordinate spaces.

[0016] FIG. 6 illustrates an example of placing AR content items within an AR environment.

[0017] FIGS. 7A-7B illustrate an example of placing a visual representation of a social-network post within an AR environment.

[0018] FIG. 8 illustrates an example method for mapping an environment with a first computing device and a second computing device.

[0019] FIG. 9 illustrates an example method for relocalizing a first computing device and a second computing device within a shared AR environment.

[0020] FIG. 10 illustrates an example of AR content items being rendered within a view of an AR environment.

[0021] FIG. 11 illustrates an example of a user post being rendered within a view of an AR environment.

[0022] FIG. 12 illustrates an example of an AR mustache being drawn on a face detected in an AR environment.

[0023] FIG. 13 illustrates an example method for rendering, on a computing device of a first user, an AR content item placed by a second user within an AR environment.

[0024] FIG. 14 illustrates an example of a view of AR environment that includes suggested AR content.

[0025] FIG. 15 illustrates an example method for suggesting an AR content item to a user.

[0026] FIG. 16 illustrates an example network environment associated with a social-networking system.

[0027] FIG. 17 illustrates an example social graph.

[0028] FIG. 18 illustrates an example computer system.

DESCRIPTION OF EXAMPLE EMBODIMENTS

[0029] In particular embodiments, an AR application may be configured to operate on any computing device, including mobile devices (e.g., smartphones, tablets, laptops, AR glasses) and other types of computing devices (e.g., desktop computers). In particular embodiments, an AR application may be configured to obtain images (e.g., video frames) of a real-world environment captured by a camera communicatively coupled to the device on which the AR application is running. By processing the images, the AR application may track real-world objects captured within the images.

[0030] In particular embodiments, the AR application may support one or more types of tracking algorithms, which may be used to create a map of a real-world environment to create an AR environment (e.g., an AR representation of the real-world environment). Users may view, interact, or otherwise engage with the AR environment on any AR compatible computing device. As an example and not by way of limitation, a user may be able to view AR content items (e.g., a doodle, a GIF, an avatar, a filter, a mask) within the AR environment. Although this disclosure focuses on AR devices and environments, it contemplates applying the same concepts in the context of virtual-reality (VR) devices and environments. For the purposes of this disclosure, the term “augmented reality” (or AR) also refers to what may sometimes be referred to as “mixed reality.”

[0031] In particular embodiments, a processor of a first computing device may receive first information from one or more sensors associated with the first computing device. This may include information about images captured at the sensors. For example, the sensors may include a camera that captures photo images. In particular embodiments, a computing system (e.g., the first computing device, a server) may receive second information from a second computing device. The second information may include information about images captured at one or more sensors associated with the second computing device. In particular embodiments, a computing system may identify one or more first points within the first images and one or more second points within the second images. These points may correspond to features of interest in the images. In particular embodiments, a computing system may “relocalize” the first computing device and the second computing device within a shared AR environment. This relocalization process may be an initialization or re-initialization process that may be used to orient an AR device within an AR environment.

[0032] The disclosed invention may provide several technological benefits. For example, the relocalization of multiple devices may allow for collaboration among multiple devices in mapping a real-world environment onto an AR environment and tracking objects within the real-world or AR environments. This collaboration among multiple devices may markedly improve the mapping process by reducing the time and effort it takes to map an environment. This improvement may be particularly pronounced and its benefits may compound significantly in cases where the real-world environment that is to be mapped is very large. As an example and not by way of limitation, the time and effort involved with mapping of a large football stadium may be significantly reduced if thirty users can relocalize and map the stadium separately to create a shared AR environment that includes all the points that they each mapped out separately.

[0033] In particular embodiments, a user may share or present an AR content item to other users within an AR environment. In particular embodiments, a first computing device associated with a first user may render an AR environment on a display associated with the first computing device. In particular embodiments, a computing system may determine that the first computing device is authorized to access a particular AR content item associated with a location within the AR environment based on social-networking information associated with the first user. The particular AR content item may be associated with a second user. In particular embodiments, the first computing device may receive information configured to render the AR content item at the associated location within the AR environment. In particular embodiments, the AR content item may be rendered on the display associated with the first computing device, such that it appears within the AR environment.

[0034] In particular embodiments, the AR content item may be integrated into the AR environment and tied to particular locations or objects identified in the environment. By so doing, a more immersive experience may be created for a user viewing an AR environment, such that the AR content may be perceived as being an integral part of the real-world environment being modeled by the AR environment. In particular embodiments, he first user may be able to view or interact with the AR content item in intuitive ways (e.g., picking them up, moving them, modifying them, etc.).

[0035] In particular embodiments, an AR environment may be rendered. In particular embodiments, one or more objects (e.g., a face, an animal, a real-world inanimate object, a virtual object, etc.) may be identified within an AR environment rendered on a display associated with a first computing device associated with a first user. In particular embodiments, a computing system may determine one or more suggested AR content items for the first user. In particular embodiments, the determination as to what content to suggest may be based on social-networking information associated with the first user. In particular embodiments, the determination as to what content to suggest may be based on characteristics associated with the objects identified in the AR environment. In particular embodiments, the suggested AR content item may be presented as suggestions on a display associated with the first computing device. A user may then be able to select a suggested AR content item and interact with it (e.g., by placing it at a location within the AR environment).

[0036] By providing a suggestion tool that intelligently determines optimal AR content that is tailored to the user and the circumstances, users may be encouraged to place AR content within an AR environment or otherwise interact with objects in the AR environment. The suggestion tool may allow for discovery of content that may be particularly relevant to a user (e.g., content created by friends of the user, content tailored to the user’s interests, etc.) at any given time.

[0037] One example of a tracking algorithm is Simultaneous Localization and Mapping (SLAM). At a high-level, SLAM is a technique that may be used to generate a 3D map of an unknown environment (e.g., in real time). The SLAM technique may employ sensor data from a variety of sensors, including, for example, cameras, LiDAR sensors, radar, gyroscope, and any other suitable types of sensors. In particular embodiments, SLAM implemented on conventional mobile phones may use the phone’s camera(s), gyroscope, and/or accelerometer. Conceptually, given a video frame, SLAM may estimate the relative position and orientation of the camera and features of interest in the scene (e.g., often edges, corners, etc.) and iteratively update the estimates based on motion and the resulting feature observations. Based on positional deviations of those features due to movement, SLAM may use triangulation techniques to generate a 3D model of the recognizable objects in the captured scene. For example, when the camera moves, a landmark point associated with a feature of interest may move. Movement of that landmark (and other landmarks) may be used to estimate the 3D position and orientation of objects.

[0038] Another example of a tracking algorithm is face tracking. In particular embodiments, face-detection algorithms may use machine-learning models to detect facial features (e.g., eyes, nose, mouth, etc.) and generate a facial mesh using points mapped to those features. In essence, a 3D model of the face may be generated to track a person’s face (and/or feature) movement and orientation. The facial mesh in particular embodiments may be an assembly of multiple (e.g., 32, 64, etc.) polygons (e.g., triangles, squares, etc.) that approximates the contours of a detected face. Using the facial mesh, the tracking algorithm may try to match the facial mesh with the face detected in each video frame. How well the facial mesh fits the face captured in a video frame may be represented as a confidence score (e.g., based on distance measurements and whether facial features corresponding to particular polygons are detectable). When the user’s face is occluded (e.g., due to another object or the face turning in a manner that hides the user’s face from view), the confidence score may drop. In particular embodiments, the confidence score may be 0 or 1, with 0 representing an undetectable face and 1 representing a detectable face (or vice versa). In particular embodiments, a simple face tracking algorithm may track a face based on estimated dimensions of a face or facial features. As an example and not by way of limitation, the simple face tracking algorithm may estimate that a face is nine inches based on a known average (which may be determined using machine-learning models that analyze images) and overlay a facial mesh that maps facial features based on average proportions of facial features.

[0039] Another example of a tracking algorithm is region tracking, which is an appearance-based tracker in accordance with particular embodiments. In particular embodiments, the region-tracking algorithm in particular embodiments mainly process 2D pixel or texture information (aside from possibly generating a plane to model the location of an object in a video frame and using gyroscope data to determine the plane’s orientation in 3D space). At a high-level, the region-tracking algorithm may seek to identify a region in a video frame that corresponds to an object and sees how it transforms in the next frame. In particular embodiments, the region-tracking algorithm may identify and track successive positions of the object in a series of video frames. Pixels of each video frame may be processed using a segmentation algorithm to identify segments that may correspond to real-world objects. The particular segmentation algorithm used may be, for example, motion-based or intensity-based, known to persons of ordinary skill in the art. Each identified segment may be labeled and represented using a polygon to approximate its shape and location within the scene. Each polygon’s motion between frames may be assumed to be a 2D affine transformation. Using the object’s polygon representation and the corresponding motion model (e.g., a machine-learning classifier) a prediction may be made as to where and how the polygon would appear in the next frame. The prediction may be compared to actual segments observed in that frame, and the object’s polygon and motion model may be updated accordingly. Over several iterations, the algorithm may be able to detect with a level of confidence that two polygons appearing in their respective frames are likely the same object and label them as such. The confidence level may drop, however, if an object is sufficiently occluded (e.g., hidden by another object), transformed beyond the predictive ability of the current model, and/or lighting conditions changed significantly to hamper image processing.

[0040] Another example tracking algorithm may simply use gyroscope data. As an example and not by way of limitation, a mobile device may have a built-in gyroscope, which may provide gyroscope data that describes the orientations of the associated mobile device. The orientation data may be purely based on the gyroscope sensor, and as such no image processing may be needed. More information on AR tracking or mapping algorithms may be found in U.S. patent application Ser. No. 15/803,428, filed 3 Nov. 2017, which is incorporated herein by reference.

Mapping and Collaboration

[0041] In particular embodiments, a processor of a first computing device may receive first information from one or more sensors associated with the first computing device. The first information may include information associated with one or more images captured at the sensors. As described elsewhere herein, the term “sensor” may include any suitable sensor (e.g., a camera, a LiDAR sensor, a radar sensor) that is capable of capturing an image. The term “image,” when used herein to describe what is captured at a sensor, refers to a representation of an environment (e.g., a representation of an environment surrounding the sensor). As an example and not by way of limitation, the term may describe a visual representation (e.g., a photo) of an environment as captured by a camera. As another example and not by way of limitation, the term may describe a collection of data representing distances to tangible surfaces of an environment as captured by a LiDAR sensor. In particular embodiments, the first information may be information sufficient to reconstruct some or all of the image. As an example and not by way of limitation, in the case where the sensor is a smartphone camera, the first information may encode an image in a raw image format (e.g., an unprocessed pixel buffer directly sourced from the camera) or any other suitable image file format (e.g., JPEG, BMP, GIF, PNG).

[0042] In particular embodiments, the first computing device may be a client computing device (e.g., a mobile computing device, a desktop device, a stylus device). In particular embodiments, the first computing device may include a processor, a sensor, and functionality for determining its location, direction, or orientation (e.g., a GPS receiver, an inertial movement unit (IMU), a compass, a gyroscope, an accelerometer). The first computing device may also include functionality for wired communication, or wireless communication, such as BLUETOOTH communication, near-field communication (NFC), infrared (IR) communication, or communication with wireless local area networks (WLANs) or cellular-telephone networks.

[0043] In particular embodiments, the first computing device may include a display that is able to render an AR environment (e.g., on an AR application). In these embodiments, the first computing device may render an AR environment that is based on the images captured at one or more sensors associated with the first computing device. As an example and not by way of limitation, the first computing device may be a smartphone with a screen that displays an AR environment based on images captured with one or more associated local cameras.

[0044] In particular embodiments, the first computing device may not include a display that is able to render an AR environment. As an example and not by way of limitation, the first computing device may be a mobile computing device that may resemble a stylus or marker (referenced herein as a “stylus device”). In this example, the first computing device may still include a sensor that captures images. As another example and not by way of limitation, the computing device may be a smartphone with a camera. In this example, the camera of the smartphone may capture one or more images that may be sent to a processor of the smartphone for processing.

[0045] FIG. 1 illustrates example embodiments of multiple devices configured to communicate information with each other. In particular embodiments, a computing system may receive second information from a second computing device. The second information may include information associated with one or more second images captured at one or more sensors associated with the second computing device. In particular embodiments, the second computing device may be a client computing device (e.g., a mobile computing device, a desktop device, a stylus device). In particular embodiments, the computing system may be the first computing device. As an example and not by way of limitation, the second computing device may be in direct wired or wireless communication with the first computing device (e.g., via BLUETOOTH). In this example, referencing FIG. 1, the first computing device 110 may receive the second information from the second computing device 120 via the connection 140. In particular embodiments, referencing the example in FIG. 1, the second computing device 120 may receive information from the first computing device 110 via the connection 140 (e.g., first information associated with first images captured at the first sensor).

[0046] In particular embodiments, the computing system may be a server computing machine. In these embodiments, the server computing machine may act as an intermediary. As an example and not by way of limitation, referencing FIG. 1, the first computing device 110 may send the first information to the server computing machine 130 via the connection 115. Similarly, in this example, the second computing device 120 may send the second information to the server computing machine 130 via the connection 125. In particular embodiments, a hybrid communication system may be user, where not all of the computing devices may be communicating with the server computing machine. As an example and not by way of limitation, referencing FIG. 1, the first communication device 110 may send the first information to the second computing device 120 via the connection 140, and the second computing device 120 may send both the first information and the second information to the server computing machine 130 via the connection 125. In this example, the computing device 110 may not have established the connection 115 (e.g., because it may lack authorization to do so, or it may otherwise be incapable of doing so for hardware or software reasons).

[0047] In particular embodiments, the first computing device and the second computing device may both be operated by a single user. As an example and not by way of limitation, the first computing device may be a device with a display (e.g., a smartphone) and the second computing device may be a device without a display (e.g., a stylus device, a controller). In this example, a user may view an AR environment with the first computing device and interact with the AR environment (e.g., by placing an AR content item in the AR environment) with the second computing device. In particular embodiments, the first computing device may be operated by a first user and the second computing device may be operated by a second user. Although this disclosure focuses on examples with two computing devices, it contemplates that any number of such devices may be in communication with each other (e.g., receiving and transmitting image information among themselves).

[0048] In particular embodiments, a computing system may identify one or more first points within the first images. In particular embodiments, a computing system may identify one or more second points within the second images. In particular embodiments, the computing system may be the first computing device or the second computing device. As an example and not by way of limitation, the first computing device (or the second computing device) may locally process the images to identify points in the images. In particular embodiments, the computing system may be a server computing machine. As an example and not by way of limitation, the computing system may be a remote server associated with the social-networking system 1660 described herein. In particular embodiments, the computing system may be a combination of multiple devices. As an example and not by way of limitation, the processing tasks required to identify points within an image may be performed by both the first computing device and a server computing machine. As another example and not by way of limitation, the processing tasks may be performed by the first computing device, the second computing device and a server computing machine.

[0049] In particular embodiments, the computing system may use one or more suitable tracking algorithms to identify these points in the first and second images. As used herein, the term “points” refers to features of interest in an image or a series of images. In particular embodiments, a point may refer to features such as corners or edges that may be identified in an image. In particular embodiments, a point may refer to an object identified in an image (e.g., a polygon corresponding to an object or face in the corresponding real-world environment). FIG. 2 illustrates an example real-world scene 200 captured within an image or video frame 205. The illustrated scene 200 includes a box 210 with a cat 220 on top. The scene 200 further includes a person sitting at a desk, with his face 230 and hand 240 visible. In particular embodiments, the video frame 205, along with successive frames, may be processed by the AR application using one or more tracking algorithms to track objects of interest. In particular embodiments, the objects of interest may include any object that a user of the AR application taps on or interacts with through the AR application (e.g., the user may associate an AR effect with a particular object, modify the object, magnify the object, etc.). In particular embodiments, the objects of interest may additionally or alternatively include any discernable or detectable object by the tracking algorithm.

[0050] FIG. 3 illustrates an example of an image with identified points associated with certain objects of interest. FIG. 4 illustrates another example of an image with identified points associated with an object of interest. The visual representation of tracking data 300 may be associated with the real-world scene 200 (e.g., an image representing at least a portion of a real-world environment) shown in FIG. 2. As an example and not by way of limitation, box 310 in FIG. 3 may represent tracking data for the real-world box 210 in FIG. 2; polygon 320 in FIG. 3 may represent tracking data for the cat 220 in FIG. 2; facial mesh 330 in FIG. 3 may represent tracking data for the person’s face 230 in FIG. 2; and circle 340 in FIG. 3 may represent tracking data for the person’s hand 240 in FIG. 2. In particular embodiments, each of these shapes and meshes may be tracked as points within the image (and ultimately, within the AR environment that may be generated based on the image). In particular embodiments, the points may be smaller visual markers within the image that track an object of interest. As an example and not by way of limitation, referencing FIG. 3, a series of visual markers 310 may track the corners of the box 310 in the image, which may correspond to the real-world box 210 in FIG. 2. In particular embodiments, the tracking data associated with the objects of interest in a scene may be generated using one or more tracking algorithms operating concurrently. For example, tracking data represented by box 310 may be generated using Simultaneous Localization and Mapping (SLAM); tracking data represented by polygon 320 may be generated using region-tracking algorithm; tracking data represented by facial mesh 330 may be generated using a face-tracking algorithm; and tracking data represented by circle 340 may be generated using a hand-tracking algorithm. In particular embodiments, the tracking algorithm used for tracking a particular object or region may be switched or adjusted independently from any other tracking algorithms used for tracking other objects or regions in the scene. For example, while the tracking algorithm used for tracking the person’s face 230 may dynamically switch from facial-tracking algorithm to region-tracking algorithm, the tracking algorithms used for tracking the other objects (e.g., the box 210, cat 220, and/or hand 240) may remain unchanged. In particular embodiments, one or more tracking algorithms may track features such as corners and edges that may correspond to real-world objects. As an example and not by way of limitation, referencing FIG. 4, a tracking algorithm may track the corners (e.g., the corner 410) of the box 420.

[0051] In particular embodiments, a computing system may relocalize the first computing device and the second computing device within a shared AR environment. As used herein, the terms “relocalizing” and “relocalization” may refer to an initialization or re-initialization process that may be used to orient an AR device within an AR environment. In cases where there is a single computing device viewing, mapping, otherwise interacting with an AR environment, relocalization may be necessary when there has been a tracking failure. As an example and not by way of limitation, when a camera of the single computing device is temporarily obscured, it may lose its bearings (e.g., location, orientation) with respect to the AR environment. In this example, the single computing device would need to be relocalized (“or re-initialized”) to the AR environment such that it understands where it is. This may be done by comparing features or points of newly acquired images with features or points in memory that were extracted from images acquired before the tracking failure. In the case of multiple computing devices (e.g., the first computing device and the second computing device), a similar relocalization process may occur as a first initialization process (or a re-initialization if a tracking failure occurs at any point). As an example and not by way of limitation, a first computing device may relocalize to a shared AR environment by comparing features or points of first images (e.g., captured by its own associated sensors) to the features or points of second images (e.g., captured by sensors of a second computing device). In this example, the first computing device and the second computing device may both orient and place each other within the shared AR environment.

[0052] FIG. 5 illustrates an example abstraction of the concept of relocalizing multiple coordinate spaces. In particular embodiments, the relocalization process may include defining a first coordinate space including the first points that were determined from the first images. As an example and not by way of limitation, referencing FIG. 5, the first coordinate space 510 may include several points (e.g., 512, 514, and 516) determined from images captured by a first computing device. These points may correspond to corners of shapes in the images (e.g., the square and the triangle in FIG. 5), or to any other suitable features (e.g., changes in texture, color, etc.). Each of these first points may be associated with a respective coordinate within the first coordinate space based on the first information. In particular embodiments, the relocalization process may include defining a second coordinate space comprising the second points that were determined from the second images. As an example and not by way of limitation, referencing FIG. 5, the second coordinate space 520 may include several points (e.g., 522, 524, and 526) determined from images captured by a first computing device. In this example, these points may correspond to corners of shapes in the images (e.g., the triangle and the circle in FIG. 5). Each of these second points may be associated with a coordinate within the second coordinate space based on the second information. In particular embodiments, the relocalization process may include identifying one or more shared points. Each shared point may be a point at which a particular first point and a particular second point intersects. As an example and not by way of limitation, referencing FIG. 5, the points 514 and 516 in the first coordinate space 510 may be determined to intersect with the points 524 and 526, respectively in the second coordinate space 520, and may consequently be identified as shared points. In particular embodiments, the relocalization process may include combining the first coordinate space with the second coordinate space to create a shared coordinate space that corresponds to a shared AR environment. This shared AR environment may be created based on the identified shared points. As an example and not by way of limitation, referencing FIG. 5, the first coordinate space 510 may be combined with the second coordinate space 520 to create the shared coordinate space 530, which may correspond to a shared AR environment. In this example, the points 514 and 524 may be determined to correspond to the same feature (e.g., the top corner of the triangle illustrated in FIG. 5) in a real-world environment, and the points 516 may be determined to correspond to the same feature in a real-world environment. As a result, the points 514 and 524 may be collapsed into a single point 534 in the shared coordinate space 530, and the points 516 and 526 may be collapsed into a single point 536 in the shared coordinate space 530. In particular embodiments, once the computing devices have been relocalized with common reference points, the locations of the computing devices (or their associated sensors) may be determined within the shared coordinate space, and thereby within the shared AR environment.

[0053] In particular embodiments, relocalization may occur among computing devices only when their respective users have authorized the pairing or the computing devices. As an example and not by way of limitation, when it is determined that a first computing device has tracked a predefined number of points in a real-world environment, and when it is determined that a second computing device is also in the real-world environment, a prompt may be provided to a first user of the first computing device to ask if the first user wants to relocalize with the second computing device (e.g., of a second user, or of the same first user). Both the first user and the second user may receive similar prompts. In this example, if both users accept, the two computing devices may relocalize, and may then collaborate in mapping the environment.

[0054] In particular embodiments, it may be determined that the first computing device and the second computing device are in the same real-world environment if they are within a threshold distance of each other. As an example and not by way of limitation, this distance may be determined by a BLUETOOTH pairing range, and the computing devices may discover each other by a BLUETOOTH discovery protocol. As another example and not by way of limitation, this distance may be a predetermined radius and the locations of the computing devices may be determined by GPS. As another example and not by way of limitation, social-networking information associated with one or more users of the computing devices may be leveraged to determine that the computing devices are within the same real-world environment. For example, if a first user enjoys pizza and dislikes burgers and if a second user has the opposite tastes, and if a pizza restaurant is next door to a burger restaurant, it may be determined that the first user is in the pizza restaurant and the second users in the burger restaurant. This may be useful in cases where the pizza restaurant and the burger restaurant are identical and very close to each other, in which case the computing devices of the two users may mistakenly relocalize within an incorrect shared AR environment. By accounting for the fact that the first user and the second user are unlikely to be in the same restaurant in this example, this mistaken relocalization may be prevented.

[0055] In particular embodiments, once multiple computing devices have been relocalized to a shared AR environment, they may each individually be aware of their location and orientation in the shared AR environment (e.g., because they may each have sensors capable of capturing image information to remain anchored within the shared AR environment). This may be advantageous in that each device may be able to reliably interact with the shared AR environment individually, without requiring corroboration from the other devices. As an example and not by way of limitation, a user may be creating an AR drawing in an AR environment by tracing its outline with a stylus device, and this process may be viewed by a tablet device that has been relocalized to the AR environment. In this example, even if the tablet device is turned away from the stylus device, the drawing process may continue uninterrupted because a camera on the stylus device ensures that the stylus remains anchored to the AR environment.

[0056] In particular embodiments, once multiple computing devices have been relocalized to a shared AR environment with the computing devices initialized to have common reference points, they may be able to collaborate to map the AR environment in real time. As an example and not by way of limitation, referencing FIG. 5, the points discovered individually by the first computing device and the second computing device may be aggregated in the shared coordinate space 530. In this example, the users may continue to map the environment by tracking additional points (e.g., points corresponding to the hexagon 540). In particular embodiments, this mapping may occur in real-time, and the computing devices that are collaborating may repeatedly sync up to share update information that indicates additional points since the last sync (e.g., newly discovered points). As an example and not by way of limitation, the computing devices may sync up periodically based on a predefined period (e.g., every two minutes) or whenever a threshold number of additional second points is identified (e.g., each time a new point is discover, each time five new points is discovered, etc.). In particular embodiments, information associated with an entire map may be sent from one computing device to another. As an example and not by way of limitation, a newly relocalized computing device may simply receive a map from another computing device (e.g., one that has the most up-to-date map).

[0057] In particular embodiments, two or more maps may be intelligently merged to create a master map. As an example and not by way of limitation, if two maps have two points (e.g., a polygon corresponding to an object) that are determined to be matching or otherwise associated with the same object or location in two maps, one of the two points may be selected for incorporation into the master map. The points may each have a confidence score that may indicate the likelihood that the point is accurate. This confidence score of a point may be determined based on, for example, the number of perspectives (e.g., as provided by different images) from which the respective point was identified. A higher number of perspectives may yield a higher confidence score for the point. For example, a point that was identified in images from ten different perspectives may have a higher confidence score than a point that was identified in images from five different perspectives. In this example, the point with the higher confidence could be incorporated into the master map. This intelligent merging may increase the accuracy of the master map. In particular embodiments, the master map may be considered the most up-to-date map, and may be sent to one or more devices (e.g., following its creation). In particular embodiments, information describing a particular point from one perspective (e.g., from a first device) may be combined with information describing the point from another perspective (e.g., from a second device). This may increase the efficiency of the mapping process and maximize the confidence of the resulting map.

[0058] In particular embodiments, any changes that occur in an environment may be continuously or semi-continuously communicated to computing devices viewing the AR environment. As an example and not by way of limitation, if a particular user moves or creates an AR content, a computing device that detects this change (e.g., the computing device of the particular user) may communicate the change to other computing devices that are viewing the AR environment. The other computing devices may accordingly update their local rendering of the AR environment and the associated local maps. As another example and not by way of limitation, if a real-world object within the environment is moved (e.g., a person may have shifted a chair), a computing device may communicate the change to other computing devices.

[0059] In particular embodiments, collaboration may be initiated only when the users involved have explicitly authorized it. As an example and not by way of limitation, users that are relocalized within an AR environment may receive a prompt asking whether they want to collaborate together in further mapping the AR environment.

[0060] In particular embodiments, a prompt for relocation or collaboration may only be offered to computing devices that are within a particular group of users. As an example and not by way of limitation, a first computing device may be in this group, with respect to a second computing device, if privacy settings associated with the second computing device explicitly include the first computing device as being part of the group. As another example and not by way of limitation, the prompt may only be offered for users who are part of the same social-network group (e.g., the group named “Cool Dog Group”) on an online social network, or users who watch tennis (e.g., as determined based on associated social-graph information). In particular embodiments, the prompt may be offered only to users who share particular characteristics (e.g., based on demographic information, based on profile information, based on user affinities as determined by social-graph information, etc.). In particular embodiments, the prompt may only be offered for users who are determined to be attending a common event (e.g., based on their location, based on their respective calendar data, based on their RSVP or registration to the event). As an example and not by way of limitation, at a hotel hosting both a medical conference and a Star Trek convention, users attending the medical conference may be prompted to collaborate with each other but not with users attending the Star Trek convention. In particular embodiments, relocalization or collaboration may only be offered for social connections. As an example and not by way of limitation, only users that are at least first-degree (or second-degree) social connections may be prompted to relocalize. In this example, an online social network may be queried (e.g., by the computing devices of the users) to determine whether users are first-degree social connections. An associated social-networking system may access a social graph and determine if nodes corresponding to the respective users are connected by a single edge (and optionally, if that edge is a friend-type edge, indicating that the users are friends on the online social network). In particular embodiments, relocation or collaboration may occur automatically (e.g., for friends of the user, for users attending the same event, for users who are part of the same group) without requiring a prompt if the associated user has explicitly allowed for this in associated privacy settings. In particular embodiments, users may override any of these restrictions and relocalize or collaborate with any users they want if all parties are in agreement.

[0061] The disclosed collaboration process may markedly improve the mapping process by reducing the time and effort it takes to map a real-world environment onto an AR environment. This improvement may be particularly pronounced and its benefits may compound significantly in cases where the real-world environment that is to be mapped is very large. As an example and not by way of limitation, the time and effort involved with mapping of a large football stadium may be significantly reduced if thirty users can relocalize and map the stadium separately to create a shared AR environment that includes all the points that they each mapped out separately. As another example and not by way of limitation, the same football stadium may be mapped by crowdsourcing information from a large number of users. For example, if a subset of attendees at a football game in the stadium have affirmatively asked or otherwise explicitly authorized a network (e.g., an online social network, a peer-to-peer network, devices of friends) to collect point data from their computing devices, their computing device may transmit the point data to the network (e.g., a server of an online social network, directly to users of a peer-to-peer network, directly to devices of friends). In particular embodiments, a real-world environment may be mapped by multiple computing devices taking photos or videos at different locations in the real-world environment if their respective users have explicitly authorized this mapping and collaboration. As an example and not by way of limitation, if a set of users attending a concert have explicitly authorized this mapping and collaboration (e.g., by going to a privacy settings menu and enabling this feature), whenever they take a photo or video with their smartphone within the concert venue, the resulting images can be used to map the environment. Alternatively, a separate sensor may be used to capture images for mapping purposes (e.g., low-resolution images captured by a different camera). In these examples, one or more images may also be captured leading up to the point where the user actually takes a photo or video (i.e., a period of time immediately preceding the taking of the photo or video as the user frames the photo or video), to further aid with tracking points and mapping the environment.

[0062] In particular embodiments, the mapping and collaboration feature may be enabled only for a period of time and may be automatically disabled after the period of time expires. In particular embodiments, the mapping and collaboration feature may be enabled only while the computing device is within a specified real-world area and may be automatically disabled after the device leaves that area. As an example and not by way of limitation, a user may enable the feature when the user is attending a concert, but the feature may be automatically disabled as soon as the user exits the concert venue.

[0063] In particular embodiments, a computing device may be used to place AR content items within an AR environment. As an example and not by way of limitation, a server (or a client computing device, for example in the case where client computing devices are in direct communication without a server as an intermediary) may receive content information from a computing device, wherein the content may include information defining one or more locations in an AR environment and information specifying an AR content item. In particular embodiments, the AR content item may be placed within the AR environment at a location associated with the defined locations. In particular embodiments, the AR environment and the AR may be rendered on a display associated with a computing device.

[0064] FIG. 6 illustrates an example of placing AR content items within an AR environment. Multiple users may collaborate to place an AR content item. In particular embodiments, “placing” the AR content item within the AR environment may include the act of drawing an AR content item. Referencing FIG. 6, as an example and not by way of limitation, a first user 610 and a second user 620 may collaborate to place a drawing 630 within the AR environment shown in a display 600 of a third user’s computing device. In this example, the first user may use a stylus device 615 and the second user may use a smartphone 625 to create the drawing 630. In this example, the first user may use a series of gestures with the stylus device 615, tracing the drawing 630 in the air with the tip of the stylus device 615. The drawing 630 may appear within the AR environment from a drawing point, which may be the tip of the stylus device 615. Similarly, the second user 620 may use the smartphone 625 to draw from the other side. In this example, the drawing point may be spaced away from the smartphone 625 by a predefined distance 640. In this example, a sensor on the computing devices (e.g., stylus device 615, the smartphone 625) may determine the locations of the computing devices while the drawing is being traced (e.g., based on a plurality of acquired images, based on a gyroscope, etc.) to determine the drawing and its location.

[0065] In particular embodiments, the AR content item may be a visual representation of a post or other communication. FIGS. 7A-7B illustrate an example of placing a visual representation of a social-network post within an AR environment. As an example and not by way of limitation, referencing FIG. 7A, a user of a first computing device such as a virtual reality device may access an interface 720 of an online social network (e.g., a newsfeed interface), which may be displayed in a portion of a view on a display unit 710 (e.g., a virtual-reality headset, and AR headset) of the first computing device. In this example the user may be able to manipulate a representation 730 of the user’s hand (e.g., with a gesture input using a controller, with a gaze input using the display unit 710) to “pick up” or otherwise select a representation of a user post 740 from the interface 720. In this example, now referencing FIG. 7B, the user may place the visual representation of the user post 740 at a specific location within the AR environment–under the painting 750 (e.g., because the user post may have been a comment from a friend about the painting 750).

[0066] In particular embodiments, the AR content item may include a photo, a video, an image file (e.g., a GIF, a meme), an emoji, a sticker, text, a filter, a mask, a drawing, or any other suitable item. In particular embodiments, the AR content item may include a virtual object. As an example and not by way of limitation, the AR content item may be a three-dimensional virtual representation of an object such as a balloon or a tree. As another example and not by way of limitation, the AR content item may be an interactive virtual object such as a video player or video game window within which a user in the AR environment can interact.

[0067] In particular embodiments, the AR content item may be placed in association with one or more detected objects in the AR environment. As an example and not by way of limitation, a user of a computing device may draw a mustache on an object that is determined to be a face of another user in the AR environment. In this example, the mustache may remain on the other user’s face indefinitely, for a defined period of time, or until some action is performed to remove the mustache (e.g., a gesture to “wash away” the mustache). In particular embodiments, the AR content item may be a mask that is configured to be overlaid on an object identified in the shared AR environment. As an example and not by way of limitation, a user of a computing device may overlay a mask on a face of another user who appears in the AR environment. The mask may be applied to a facial mesh of the face, such that in the AR environment, it may appear that the face has a mask over it. The mask may cover the entire face or just a portion of it (e.g., it may simply be a mustache mask). In particular embodiments, the AR content item may be a filter overlay that is configured to be overlaid on a rendering of the shared AR environment. As an example and not by way of limitation, a user may apply a Holiday-themed filter to an environment, which may automatically add virtual objects (e.g., a virtual wreath, virtual snow) to locations within the AR environment, or affect color, brightness, contrast, or other suitable display settings.

发表评论

电子邮件地址不会被公开。 必填项已用*标注