Abstract: System and method to perform targeted image adjustments starts with a processor receiving a media content item and identifies an image adjustment parameter and an adjustment value based on the media content item. Processor generates an adjusted media content item using the image adjustment parameter and the adjustment value and causes an adjustment interface to be displayed by a display of a user device. The adjustment interface can comprise the adjusted media content item and a selectable item associated with the image adjustment parameter. The selectable item can include settings. In response to receiving a selection of one of the settings of the selectable item, processor generates a final media content item based on the selection of the one of the settings, and cause the final media content item to be displayed by the display of the user device. Other embodiments are described herein.

Abstract: Technological improvements are described for predicting user engagement with newly published, un-viewed, relatively short videos based on features extracted from the video content itself. A network includes an extraction module for extracting features from the video content and encoding the text-based features. The visual features are processed using learnable, per-frame and per-clip multi-layer perceptrons (MLPs) and using multi-modal cross-attention mechanisms. A fusion module merges a set of predicted features based on the results generated by the MLPs. A temporal aggregation module uses a multi-layer self-attention architecture to combine the fused features and obtain a set of temporal aggregated features. The network uses two branches (consisting of MLP layers) to jointly predict two new engagement metrics: a normalized action watch percentage (NAWP) and an engagement continuation rate (ECR).

Abstract: Embodiments of the present disclosure relate generally to systems for presenting content suggestions within a messaging interface. A content suggestion system performs operations that include: receiving inputs that include a text string into a messaging interface; parsing a keyword from the text string of the input in response to receiving the inputs; retrieving message content that includes at least a media item based on the keyword; and presenting a chat suggestion that includes the message content within the messaging interface, according to certain example embodiments.

Abstract: A wearable augmented reality device enables rapid, personalized messaging through a body-anchored communication interface. The device captures image data via cameras, detects and tracks a body part of the wearer, and displays a user interface anchored to the detected body part. The interface comprises user interface elements representing social connections. Upon detecting selection of an element, the device activates a microphone to record an audio message for delivery to the corresponding social connection(s). The interface may be dynamically updated to reflect changes in relationship attributes, messaging activity, or connection status. User interface elements can be ordered based on social connection scores or communication recency, with additional graphical elements indicating device types or custom avatars. This system facilitates seamless integration of digital communication with the physical world, enhancing social media and instant messaging experiences in mixed reality environments.

Abstract: Among other things, embodiments of the present disclosure improve the functionality of electronic messaging software and systems by generating and selecting customized media content items (such as images) with avatars of different users within electronic messages based on the context of communications between the users. For example, users of different mobile computing devices can exchange electronic communications with each other, and the system can analyze these communications to present options for media content items containing the users' avatars based on content in the communications, actions or events taken by or involving the users, or combinations thereof. The users may select such media content items for inclusion in their electronic communications.

Abstract: Methods and systems are disclosed for performing operations for estimating power usage of an AR experience. The operations include: accessing resource utilization data associated with execution of an augmented reality (AR) experience; applying a machine learning technique to the resource utilization data to estimate power consumption of the AR experience, the machine learning technique being trained to establish a relationship between a plurality of training resource utilization data associated with training AR experiences and corresponding ground-truth power consumption of the training AR experiences; and adjusting one or more operations of the AR experience to reduce power consumption based on the estimated power consumption of the AR experience.

Abstract: Systems and methods are provided for determining that a message received from a second computing device associated with a second user is an interactive message and causing the interactive message to be displayed. In response to a selection to skip engagement with the interactive message, the systems and methods display a message feed with a list of a plurality of users and cause a selectable engagement indicium to be displayed next to a list entry of the second user in the list. The system and methods further determine that a predetermined amount of time has elapsed since the start time of the display of the interactive message without engagement with the interactive message and cause the interactive message to be deleted from the message feed and the selectable engagement indicium removed from next to the list entry of the second user.

Abstract: Aspects of the present disclosure involve a system comprising a computer-readable storage medium storing a program and method for updating avatar clothing of a user of a messaging system. The program and method provide for receiving, by a messaging application running on a device of a user, a request to identify at least one object depicted in an image captured by a device camera; determining, in response to receiving the request, an attribute of an object depicted in the image; selecting, from plural candidate articles of clothing for an avatar of the user, an article of clothing that is associated with the attribute of the object; and displaying the selected article of clothing with the avatar of the user.

Abstract: Methods and systems are disclosed for suggesting scenarios for an image using one or more machine learning models based on a detected event. The methods and systems detect, by an interaction system associated with a first user, an event associated with a second user and generate a prompt comprising the event and a request for a plurality of scenarios that are relevant to the event. The methods and systems process the prompt by a large language model (LLM) to generate the plurality of scenarios that are relevant to the event and present an individual content item corresponding to an individual scenario of the plurality of scenarios.

Abstract: A system and method for generating contextually relevant reply suggestions for media posts is disclosed. The system analyzes media content to identify visual objects, scenes, text, and metadata attributes using computer vision techniques. Identified objects and attributes are incorporated into structured prompt templates to construct detailed natural language descriptions of the media context. The prompts are provided to a text generation artificial intelligence (AI) that outputs a plurality of contextual reply suggestions based on the media analysis. Suggestions are displayed as selectable options adjacent to the media post. Users can cycle through suggestions and select a reply to send. Selections are logged to improve the AI model. Feedback on suggestion quality can also be collected. By integrating computer vision and AI generation driven by engineered prompts, the system produces highly relevant, personalized responses tailored to media content.

Abstract: A system and method provide feedback to a user, such as a visually impaired user, to guide the user to an object in the field of view of a camera mounted on a frame worn on the head of the user. A processor identifies at least one object and a body part of the user in the field of view of the camera and tracks relative positions of the body part relative to the identified object. The processor also generates and communicates at least one control signal for guiding the body part of the user to the identified object to a user feedback device worn on or adjacent the body part of the user. The feedback device receives the control signal(s) and converts the control signal(s) into at least one of sounds or haptic feedback that guides the body part to the identified object.

Abstract: An eyewear device with flexible frame for Augmented Reality (AR) is disclosed. At least two sensors and a display are mounted on the flexible frame. When in use, the real time geometry of the eyewear device may change from factory calibrated geometry, resulting in low quality AR rendering. A modeling module is provided to model the real time geometry of the eyewear device on the fly using sensor information of the at least two sensors. The modeled real time geometry is then provided to a rendering module to accurately display the AR to the user.

Abstract: Aspects of the present disclosure involve a system comprising a computer-readable storage medium storing a program and method for presenting change of presence sounds within a virtual working environment. The program and method provide for receiving, from a first client device of a first participant of a group of participants within a virtual working environment, interaction data relating to the first participant being present at the first client device, the group of participants having been selected from among plural participants of the virtual working environment; determining a change in presence of the first participant with respect to the first client device; generating, in response to the determining, a sound corresponding to the change in presence; and providing the generated sound to one or more second client devices of respective one or more second participants of the group of participants, for presentation on the one or more second client devices.

Abstract: Aspects of the present disclosure involve a system comprising a storage medium storing a program and method for providing augmented reality content in association with a live event. The program and method provide for accessing first live video provided to a media server, the first live video having been captured by a camera in association with a live event; accessing timeline data stored by the media server, the timeline data for synchronizing effects with respect to the first live video; causing, based on the timeline data, an output device to display the first live video with preselected augmented reality content; and providing, to a plurality of client devices, an indication of the timeline data, each client device being configured to capture respective second live video and to display the respective second live video together with the augmented reality content based on the indication of the timeline data.

Abstract: The technical problem of matching records in different datasets, for example a host dataset and a partner dataset storing records representing respective users, while maintaining the privacy of each dataset, is addressed by providing a privacy safe joint identification protocol. The privacy safe joint identification protocol computes respective anonymous joint identifiers for records in the two datasets. An anonymous joint identifier is generated such that the host-assigned and the partner-assigned identifies that have been determined to represent the same user are mapped to the same anonymous joint identifier.

Abstract: A user provided the opportunity to unlock an extended reality (XR) wearable device. Example methods include receiving a wakeup signal, capturing an image using an image capturing device of the XR wearable device, identifying a body part of a user within the image, verifying the body part is a body part of an authorized user, and in response to a determination that the body part is the body part of the user, unlocking the XR wearable device. The method may include capturing a plurality of images using the image capturing device, and identifying a gesture performed by the user.

Abstract: A support arm assembly for a head-worn device provides radio frequency (RF) shielding for a projector. A metal support arm, configured to structurally attach to a rear structural element and an optical element holder of the head-worn device, forms a rear face, a bottom face, and a top face of an enclosure. A metal front face of the enclosure attaches to the optical element holder, and defines a front aperture for permitting passage of light from an exit pupil of the projector toward an input optical element. The metal support arm forms a structural support joining the optical element holder to the rear structural element without placing mechanical load on the projector. A first side face of the enclosure and a second side face of the enclosure are electrically coupled to the metal support arm.

Abstract: A method for aligning coordinate systems of user devices in an augmented reality system using somatic points of a user's hand as alignment markers. Images captured from multiple user devices are used to align the reference coordinate systems of the user devices to a common reference coordinate system. In some examples, user devices capture images of a hand of a user and use object recognition to identify somatic points as alignment markers. The somatic points of a user device are translated to a common reference coordinate system determined by another user device.

Abstract: The subject technology receives frames of a source media content, the frames of the source media content including representations of a head and a face of a source actor. The subject technology generates sets of source pose parameters. The subject technology receives at least one target image, the at least one target image including representations of a target head and a target face of a target entity. The subject technology generates, based at least in part on the sets of source pose parameters, an output media content, each frame of the output media content includes an image of the target face in at least one frame of the output media content. The subject technology provides an online advertisement based at least in part on the output media content for display on a computing device.

Abstract: Eyewear devices that include two SoCs that share processing workload. Instead of using a single SoC located either on the left or right side of the eyewear devices, the two SoCs have similar assigned responsibilities to operate the same types of peripheral components and perform similar processes to balance workload. In one example, the eyewear device utilizes a first SoC to operate a first color camera, a first computer vision (CV) camera, and a first display, and perform three-dimensional graphics and compositing. A second SoC operates a second color camera, a second CV camera, and a second display. The SoCs are synchronized since peripheral components are operated by both SoCs.