Abstract: In one embodiment, a method includes determining, by an application running on a client system of a user, one of a plurality of cameras of the client system being used to capture an image. The plurality of cameras may include a front camera and a back camera of the client system. The application may identify a user intent with respect to capturing the image based on one or more user-intent indicators. The user intent may indicate a type of image that the user is intending to capture. Responsive to identifying the user intent, the image may be captured using the determined camera.

Abstract: In one embodiment, a method includes determining, by an application running on a client system of a first user, one of a plurality of cameras of the client system being used to capture an image. The plurality of cameras may include a front camera and a back camera of the client system. The application may detect one or more of the first user and one or more second users using the determined camera. The application may determine whether one or more image-capturing criteria are satisfied. If the one or more image-capturing criteria are satisfied, the application may capture the image using the determined camera. Otherwise, the application may provide feedback to one or more of the first user and the one or more second users to perform one or more actions in order to comply with the image-capturing criteria.

Abstract: The present disclosure is directed toward systems and methods for interacting with portions of digital video within a social networking system. For example, systems and methods described herein enable a user to select a portion of a digital video and share or like just the portion of the digital video, rather than the entire digital video. The present disclosure is also directed toward systems and methods for identifying viral portions of a digital video within a social networking system. For example, systems and methods described herein analyze social networking activity related to portions of a particular digital video to identify one or more viral portions of the digital video. In response to identifying one or more viral portions, systems and methods provide an indication of, or otherwise share, the one or more viral portions during playback of the digital video.

Abstract: In one embodiment, a server computing machine provides an interface operable to permit a first user to broadcast a live video to one or more second users. The server computing machine provides sponsored content items to the second users while also providing the live video. In response to receiving an indication that the first user selected an option to display a sponsored content item while broadcasting the live video, the server computing machine reduces the image size of the live video, superimposes the live video of reduced size on the sponsored content item to define a combined image having a sponsored portion on which the sponsored content item is displayed and a live portion on which the live video is displayed. The combined image is then sent to the one or more second users.

Abstract: In one embodiment, a server computing machine provides an interface operable to permit a first user to broadcast a live video to one or more second users. The server computing machine selectively provides customized sponsored content items to the second users while pausing the broadcast of the live video. In response to receiving an indication that the first user selected an option to display a sponsored content item while broadcasting the live video, the server computing machine begins loading sponsored content items to the second users while the live video is still being broadcasted. After the sponsored content items have been loaded, the first and second users may be alerted prior to substantially simultaneously beginning to display the loaded sponsored content items to the second users.

Abstract: In one embodiment, a method includes displaying a feed of an online social network on a client computing device of a user, the feed comprising a plurality of media items; receiving a first indication the user is viewing a first media item, wherein the first media item is located at a first location in the feed; sending a first sponsored content item for playing at the first location; receiving a second indication the user is transitioning from the first location to a second location in the feed; and pausing the first sponsored content item from playing in response to receiving the second indication.

Abstract: In one embodiment, a method includes sending a media item to a client computing device of a user; determining an interest level of the user for the media item, wherein the interest level is determined based on a duration of time for which the media item is played on the client computing device; and if the interest level of the user is greater than a threshold interest level, then sending, to the client computing device, a sponsored-content indicator indicating that a sponsored content item will be presented and causing the sponsored content item to be presented on the client computing device.

Abstract: Systems, methods, and non-transitory computer readable media can identify an object included in a video content item. A content item associated with the object can be determined during playback of the video content item, wherein the content item is one or more of: a two-dimensional content item or a three-dimensional content item. An indicator associated with the content item can be provided in the video content item. The content item can be provided in response to a selection of the indicator.

Abstract: In one embodiment, a method includes sending, to a social-networking system, a request to access a live-broadcast session associated with a broadcasting user of the social-networking system. The live-broadcast session includes a media stream, where the media stream is available in a video format or an audio format. The method also includes determining a format for the media stream based at least in part on availability of the media stream in the video format or the audio format and a network connectivity status associated with the client system. The method further includes receiving the media stream in the determined format.

Abstract: In one embodiment, a method includes sending, to a social-networking system, information to initiate a live-broadcast session for distribution of a media stream using the social-networking system. The method also includes capturing an incoming media stream in association with the live-broadcast session, generating an outgoing media stream based on the incoming media stream, and sending the outgoing media stream to the social-networking system. The method also includes providing information to display a user interface associated with the live-broadcast session, where the user interface includes a visual representation associated with the outgoing media stream.

Abstract: In one embodiment, a method includes retrieving, from one or more data stores, a script including multiple text strings, where the script is associated with a user of a social-networking system. The method also includes capturing an incoming media stream including audio data corresponding to vocal expression by the user, where the media stream is transmitted to the social-networking system for broadcast and identifying, using a speech recognition process, one or more words in the vocal expression corresponding to a text string of the script. The method also includes providing the corresponding text string for display in conjunction with a subsequent text string of the script.

Abstract: Systems, methods, and non-transitory computer readable media can provide a virtual content item relating to a virtual three-dimensional space. An object within the virtual content item can be determined. One or more content items to provide within the virtual content item in association with the object can be determined. A three-dimensional position of the one or more content items within the virtual three-dimensional space of the virtual content item can be determined.

Abstract: The present disclosure is directed toward systems and methods for interacting with portions of digital video within a social networking system. For example, systems and methods described herein enable a user to select a portion of a digital video and share or like just the portion of the digital video, rather than the entire digital video. The present disclosure is also directed toward systems and methods for identifying viral portions of a digital video within a social networking system. For example, systems and methods described herein analyze social networking activity related to portions of a particular digital video to identify one or more viral portions of the digital video. In response to identifying one or more viral portions, systems and methods provide an indication of, or otherwise share, the one or more viral portions during playback of the digital video.

Abstract: The present disclosure is directed toward systems and methods for interacting with portions of digital video within a social networking system. For example, systems and methods described herein enable a user to select a portion of a digital video and share or like just the portion of the digital video, rather than the entire digital video. The present disclosure is also directed toward systems and methods for identifying viral portions of a digital video within a social networking system. For example, systems and methods described herein analyze social networking activity related to portions of a particular digital video to identify one or more viral portions of the digital video. In response to identifying one or more viral portions, systems and methods provide an indication of, or otherwise share, the one or more viral portions during playback of the digital video.

Abstract: A method is provided including outputting, by a communications terminal, a first voice message identifying a first screen that is displayed on a touchscreen of the communications terminal; and executing a command in response to a touch input being received at the touchscreen.

Abstract: Embodiments include a method of providing an assistant application that identifies a plurality of applications available to a device and receiving, via the device, natural language input. The natural language input can be evaluated to identify a subset of the plurality of applications in order to provide output comprising one or more suggested commands. Each suggested command can correspond to one of the subset of identified applications. In response to selection of a suggested command, the corresponding application can be invoked. Prior to invoking the application, the context for invoking the application and/or the context of the input may be evaluated in order to determine one or more parameters associated with the application. The natural language input can be used to suggest commands that include one or more suggested parameter values to pass to when invoking the application. Similar techniques can be used for accessing data services.