Abstract: An optical system for restoring a natural image combined with a digital image, in order to characterise and highlight the objects represented on the natural image. The optical system includes an objective lens, an eyepiece, a semi-reflective plate, a processing unit, capturing capabilities and restoring capabilities. Other embodiments relate to a method for producing such a digital image.

Abstract: The disclosed computer-implemented method for restricting capture of self-portrait images on mobile devices may include (i) capturing sensor data associated with a surrounding environment, (ii) detecting a user input for taking one or more self-portrait images, determining, based on the sensor data, one or more potential hazards associated with the taking of the self-portrait images within the surrounding environment, and (iv) performing a security action to protect against the potential hazards associated with the taking of the self-portrait images within the surrounding environment. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: An example system includes an electronic device and a server. The electronic device includes a display; a first communicator; and a first controller configured to: control to transmit, to the server, location data identifying locations of the electronic device. The server includes a second communicator; and a second controller configured to: control to receive the location data; identify a course of travel; link photograph images to locations on the course of travel; and control to transmit to the electronic device the course of travel and the photograph images. The first controller of the electronic device is configured to control to receive the course of travel and the photograph images and to control to display on the display, at the same time, a map image, the course of travel, and one or more of the photograph images linked to a selected location on the course of travel.

Abstract: Systems and methods are provided for retail environments. In one implementation, a non transitory computer-readable medium may include instructions that when executed by a processor cause the processor to perform a method. The method may include receiving camera location information associated with a plurality of cameras deployed in a retail store, the camera location information being indicative of an actual position of each camera of the plurality of cameras; receiving, for each camera of the plurality of cameras, at least one camera status indicator; using the received camera location information and the received camera status indicators to generate a digital signal; and providing the generated digital signal to an extended reality device to thereby cause the extended reality device to display a visual representation of the at least one status indicator for each camera of the plurality of cameras relative to the actual position of the camera in the retail store.

Abstract: An automatic emergency reporting system for a vehicle includes a determiner, a normality transmitter, a detector, and an emergency transmitter. The determiner determines a quality of communication between a communication terminal of the vehicle and a server. The normality transmitter transmits rich information available in the server from the communication terminal in a normal state. The detector detects an emergency of the vehicle. The emergency transmitter transmits emergency information from the communication terminal to the server after the detection. The emergency transmitter transmits the rich information with the emergency information in response to the detection of the emergency of the vehicle in a state in which a status of the communication with the server is satisfactory. The emergency transmitter transmits the emergency information in response to the detection of the emergency of the vehicle in a state in which the status of the communication with the server is not satisfactory.

Abstract: The present invention provides a system and methods for notifying first responders of the general or specific location of a security crisis or threat in a building or public location, and the type of threat or crisis that has occurred, while notifying building occupants or others in the public location of the crisis and how to respond. The crisis management and notification system provides critical information to the first responders, including initial location of the crisis and whether the crisis location has changed in real time, audio and video input of the crisis arena, communications with designated occupants in the crisis arena, static building or location information, and other information. The crisis notification system can be scaled to allow the effective use in facilities of differing sizes and layouts. The system is also flexible, enabling the system to integrate with currently existing systems or to operate with new devices.

Abstract: This disclosure describes a system for presenting items to a user at a presentation area within a materials handling facility. In some instances, a predicted items list that identifies items that are likely to be picked by a user are determined and, when the user arrives at the materials handling facility, those predicted items are presented to the user for selection. For example, predicted items may be determined and inventory holders that contain the predicted items may be routed to a presentation area and positioned for presentation to the user. The user may browse the presented items and pick the items they desire.

Abstract: Provided is a tangible, non-transitory, machine-readable medium storing instructions that when executed by a processor effectuate operations including: obtaining, with one or more rangefinder sensors positioned on a mobile automated device, distances from the one or more rangefinder sensors to a surface; monitoring, with the processor, the distances sensed by each of the one or more rangefinder sensors; detecting, with the processor, an edge when a change in the distances is greater than a predetermined amount; and actuating, with the processor, the mobile automated device to execute one or more movement patterns upon detecting the edge, wherein the one or more movement patterns initiates movement of the mobile automated device away from the area where the edge was detected.

Abstract: In various examples, cloud computing systems may store frames of video streams and metadata generated from the frames in separate data stores, with each type of data being indexed using shared timestamps. Thus, the frames of a video stream may be stored and/or processed and corresponding metadata of the frames may be stored and/or generated across any number of devices of the cloud computing system (e.g., edge and/or core devices) while being linked by the timestamps. A client device may provide a request or query to dynamically annotate the video stream using a particular subset of the metadata. In processing the request or query, the timestamps may be used to retrieve video data representing frames of the video stream and metadata extracted from those frames across the data stores. The retrieved metadata and video data may be used to annotate the frames for display on the client device.

Abstract: Methods, apparatuses, and non-transitory machine-readable media associated with data transmission are described. Data transmission and remote activity monitoring can include detecting a triggering event and determining an output data type associated with the triggering event, wherein the output data type is a first type for display at a second device or a second type to initiate communication between the first device and one or more second devices. Data transmission can include transmitting the output data to the second device via a device-to-device data link in response to determining the output data type comprises the first type. In response to determining the output data type comprises the second type, data transmission can include initiating a two-way communication path with the second device, the two-way communication path comprising a device-to-device data link or a data link with a base station or access point and transmitting the output data to the second device.

Abstract: An electronic device is disclosed. In addition, various embodiments identified through the specification are possible. The electronic device includes a display, a processor, and a memory storing instructions that, when executed by the processor, cause the processor to display, when a video supporting a plurality of orientation regions is played, a first screen corresponding to a first orientation region among the plurality of orientation regions, display a timeline representing a playback time of the video, display thumbnails of screens corresponding to the plurality of orientation regions at the first time point in response to a first user input pointing a first time point in the timeline, and receive a second user input scrolling the thumbnails.

Abstract: A method for determining one or more vital signs of a person includes recording video images of a scene with an egocentric camera coupled to the person's body, detecting and magnifying image frame-to-image frame movements in the video images of the scene, representing the magnified image frame-to-image frame movements in the video images of the scene by a one-dimensional (1D) amplitude-versus-time series, and transforming the 1D amplitude-versus-time series representation into a frequency spectrum. The method further includes identifying one or more local frequency maxima in the frequency spectrum as corresponding to one or more vital signs of the person.

Abstract: Techniques for updating a position of overlaid image content using IMU data to reflect subsequent changes in camera positions to minimize latency effects are disclosed. A “system camera” refers to an integrated camera that is a part of an HMD. An “external camera” is a camera that is separated from the HMD. The system camera and the external camera generate images. These images are overlaid on one another and aligned to form an overlaid image. Content from the external camera image is surrounded by a bounding element in the overlaid image. IMU data associated with both the system camera and the external camera is obtained. Based on that IMU data, an amount of movement that the system camera and/or the external camera have moved since the images were originally generated is determined. Based on that movement, the bounding element is shifted to a new position in the overlaid image.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a storage device, are disclosed. A system includes one or more processors and one or more computer storage media storing instructions that are operable, when executed by the one or more processors, to cause the one or more processors to perform operations. The operations include: obtaining an image from a camera coupled to a garage; determining that a garage door has opened; in response to determining that the garage door has opened, determining that a vehicle position satisfies an alert criteria based on the image from the camera; and in response to determining that the vehicle position satisfies the alert criteria based on the image from the camera, providing a visual alert for a driver of the vehicle on a signaling device coupled to the garage.

Abstract: A recording device captures information and stores the information in a data store as a data record. The recording device includes a low-power wireless interface and a high-power wireless interface. A network device includes a low-power wireless interface and a high-power wireless interface. An activation device transmits a data notification and a network notification. Upon receiving the data notification, the network device activates its high-power wireless interface. Upon receiving the network notification, the recording device activates its high-power wireless interface. The recording device transmits the data record to the network device via its high-power wireless interface.

Abstract: Time-stamped images are captured for unknown items within a store. Portions of the images are cropped to restrict the cropped images to attributes associated with the unknown items. A set of time ordered cropped images are proved to a trained machine-learning item recognition algorithm, which produces an output of confidence values that map to specific items of the store. When the confidence value meets or exceeds a predefined threshold, the specific item associated with that confidence value is used to identify the unknown item. The known item is assigned to an appropriate shopping cart of a shopper based on tracking from the images of the shopper while in the store.

Abstract: A device for interfacing with an interactive program is provided, including: a housing, including, an IR projector for projecting an IR pattern into an interactive environment; a camera for capturing images of the projected IR pattern; an image processor configured to process the captured images for identification of the projected IR pattern; a transceiver configured to communicate with a computing device, the transceiver configured to transmit data generated by the image processor to the computing device, such that a position of the device in the interactive environment is determined based on the captured images of the IR pattern, the determined position of the device being utilized as input for the interactive program.

Abstract: A recording system comprises an auxiliary recording device with a first camera and a (primary) recording device with a second camera. The recording device is configured to determine an activation event and to instruct at least one of the first camera to capture a first image or a second camera to capture a second image. Responsive to an image being captured, the recording device receives and stores the captured image. The recording system may comprise a body-worn battery configured to provide power to the recording system. The recording system may comprise a detachable body configured to detachably couple to a main housing body, wherein a camera of the detachable body is configured to capture images when coupled or when detached.

Abstract: A surveillance system includes a first electronic device, a second electronic device, an image-capturing unit, and a multi-computer switching unit. The image-capturing unit corresponds to the second electronic device, and the image-capturing unit captures an image of a location of the second electronic device. The multi-computer switching unit is coupled to the first electronic device, the second electronic device, and the image-capturing unit, and the multi-computer switching unit receives the image from the image-capturing unit. When the second electronic device sends a connection request for establishing a communication connection with the first electronic device, the communication connection is established through the multi-computer switching unit, and the multi-computer switching unit outputs the image corresponding to the location of the second electronic device to a surveillance terminal device.

Abstract: A server for managing inspection information of one or more structures includes circuitry configured to store identification information uniquely assigned to an information source associated with a structure, and image data of the structure previously captured in association with each other as an inspection result of the structure; receive from a communication terminal, via a communication network, the identification information of the information source acquired by the communication terminal; create data of a screen used for displaying the image data of the structure previously captured corresponding to the identification information received from the communication terminal; and transmit the data of the screen used for displaying the image data of the structure previously captured to the communication terminal via the communication network.

Abstract: Systems and methods for using radio frequency signals and sensors to monitor environments (e.g., indoor building and adjacent outdoor environments) are disclosed herein. In one embodiment, a system for providing a wireless asymmetric network comprises a hub having one or more processing units and at least one antenna for transmitting and receiving radio frequency (RF) communications in the wireless asymmetric network and a plurality of sensor nodes each having a wireless device with a transmitter and a receiver to enable bi-directional RF communications with the hub in the wireless asymmetric network. The one or more processing units of the hub are configured to determine localization of the plurality of sensor nodes within the wireless asymmetric network, to monitor loading zones and adjacent regions within a building based on receiving information from at least two sensor nodes, and to determine for each loading zone whether a vehicle currently occupies the loading zone.

Abstract: A collaborative video camera system of an edge node can be used to track and predict objects, locations of the objects, and events associated therewith. For example, multiple cameras can be utilized to determine the direction in which an object is heading. This data can be used to activate and/or dispatch other cameras that may be at or near the predicted location of the object. Additionally, sound associated with the object can be used to predict and/or active cameras that are at or near the predicted location of the object.

Abstract: Optical adaptive viewport display systems and methods are provided. One such optical adaptive viewport display system has an adaptive pupil device which is optical coupled to an optical combiner. The adaptive pupil device is optically couplable to an image projector and is configured to select a sub-pupil from the pupil of the projector. The selected sub-pupil is optically relayed by relay optics from the adaptive pupil device to an eyebox. The relay optics includes an optical combiner. The sub-pupil size and position is selected by the adaptive pupil device so that an optical image spot beam from the sub-pupil and reflected by the optical combiner on to the eye box has a diameter at the eyebox such that the virtual image, as seen by a human eye disposed at the eyebox, is hyperfocused.

Abstract: An artificial reality system can display notifications in various configurations including in a minimized version or a maximized version. The artificial reality system can initially display a minimized version of the notification. Upon identifying a first trigger event, such as the user's gaze resting on the minimized version of the notification or a rotation of a user's wrist, the artificial reality system can change the display of the notification from the minimized version to the maximized version. The artificial reality system can identify a second trigger event, such as the user's gaze not being directed to the maximized version of the notification or a rotation of the wrist in an opposite direction from the first wrist rotation. In response to the second trigger event, the artificial reality system can stop display of the maximized version of the notification.

Abstract: A wristwatch band for a computerized watch base has a bracelet portion and a connector element. The bracelet portion extends in a length direction at least partially around a wrist of a wearer and has a power supply device to supply power to the computerized watch base. The connector element secures the computerized watch base to the bracelet portion. In the length direction, the bracelet portion may be discontinuous and terminate at first and second clasp ends. First and second clasp sections may be provided respectively at the first and second clasp ends of the bracelet portion, the first and second clasp sections releaseably connecting around the wrist of the wearer. Electrical connectors may be provided in each clasp section such that when the clasp is connected around the wrist of the wearer, an electrical connection is made from the first clasp end to the second clasp end.

Abstract: Example implementations described herein are directed to a solution to the problem of accurate real-time inventory counting and industrial inspection. The solution, involves a device such as a mobile device that assists a human operator on the field to quickly achieve high quality inspection results. The example implementations detect objects of interest in individual image snapshots and use location and orientation sensors to integrate the snapshots to reconstruct a more accurate virtual representation of the inspection area. This representation can then be reorganized in various ways to derive inventory counts and other information that are not planned originally.

Abstract: A system and method that recognizes and highlights objects in a video frame for augmented video production workflow. Ae video production camera is provided to capture a video frame and generate frame data and object data relating to objects within the video frame that are in focus. Moreover, a display device is included that is wearable by the camera operator and includes a camera that captures field of view data that corresponds to a field of view of the camera operator. Furthermore, the display device can compare the video frame data with the field of view data captured by the camera to generate visual indicators for the video frame and the in focus objects. As a result, the display device can provide the camera operator with a field of view having visual indicators of the video frame and the objects that appears as overlays over the operator's field of view.

Abstract: A method and system for online decision making of generator start-up, determining the lines to be restored step by step based on the real-time power system data to guide the initial stage. The black-start unit is determined, and the units waiting to be restored are selected from all units. The training set including labeled samples to establish a value network. During generator start-up, the blackout area is judged, the state of power system is identified, the availability of equipment is judged, and the characteristics of generators are collected. The total generation capability is used as search objective for Monte Carlo tree search. Based on the value network, Monte Carlo is applied to search the next line to be restored during generator start-up. Parallel computation adopted to check the constraints for the restoration of alternative lines. The results of Monte Carlo are summarized to determine the line to be restored next.

Abstract: Eyewear is provided including a frame, and a camera connected with the frame, in which the camera is configured to be controlled by a remote controller. The camera may be configured to capture video and/or a photo. The eyewear may include data storage, and the camera may be connected to the data storage. A wrist watch may be configured to act both as a time piece and a controller of the camera. The eyewear may also include a heads up display and/or a video file player. The eyewear may also include an electro-active lens.

Abstract: Provided is an image processing apparatus that processes a video, including continuous images expressed by a coordinate system which includes at least an angular coordinate around a predetermined axis. The image processing apparatus includes: video data acquisition means that acquires video data; sensor data acquisition means that acquires sensor data corresponding to the video data; and rotation correction means that corrects the image to cancel rotation change around the reference axis based on the sensor data so that a reference of display of the image is fixed to a specific direction during the image capturing throughout a plurality of frames of the video.

Abstract: A system comprising a scanning device and a physical scannable element. The scannable element is optically captured by the scanning device and associated with a two-dimensional code. The two-dimensional code is then decoded and translated into a vector to identify active data stored on a server and the vector is sent to the server via a suitable network and active data is pushed to the scanning device.

Abstract: A two-piece headset with audio for AR (Augmented Reality)/VR (Virtual Reality)/MR (Mixed Reality), including a headpiece that is readily cleaned and disinfected, that in use releasably couples with an audio-visual component, so as to avoid physical contact between the audio-visual component and the user's head, and featuring automatic disengagement of headphones on decoupling of the headpiece and audio-visual component. An amusement attraction using two-piece headset with audio and having a support and retraction assembly for the audio-visual component.

Abstract: Systems and methods disclosed herein relate to assigning dynamic eye-gaze dwell-times. Dynamic dwell-times may be tailored to the individual user. For example, a dynamic dwell-time system may be configured to receive data from the user, such as the duration of time the user takes to execute certain actions within applications (e.g., read a word suggestion before actually selecting it). The dynamic dwell-time system may also prevent users from making unintended selections by providing different dwell times for different buttons. Specifically, on a user interface, longer dwell times may be established for the critical keys (e.g., “close” program key, “send” key, word suggestions, and the like) and shorter dwell times may be established for the less critical keys (e.g., individual character keys on a virtual keyboard, spacebar, backspace, and the like).

Abstract: Methods and apparatus, including computer program products, implementing and using techniques providing privacy masks on a video stream. In a video received from a monitoring camera, at least one image region is selected, to which region a privacy mask should be applied. For any image frame in the video that contains the at least one image region, a flag is set to indicate a non-displayed image frame. A forward-predicted image frame referencing the non-displayed image frame is inserted, wherein coding units representing the at least one image region in the forward-predicted image frame collectively provide the privacy mask by obfuscating the at least one image region.

Abstract: An apparatus for screening, treatment, monitoring and/or assessment of visual impairments, comprising electronic means for simultaneously applying two separate and unrelated processing methods to images presented to a patient's eyes: a first processing method being applied to an non-amblyopic eye (the eye with the better vision), and a second processing method being applied to an amblyopic eye (the weaker eye, or the impaired eye). A method for screening, treatment, monitoring and/or assessment of visual impairments, comprising: a. defining a starting point, wherein differences between a patient's eyes are completely, or as closely as practically possible, corrected, to enable two identical or similar images to be transferred to the brain from the patient's eyes; b. defining an ending point, wherein there is no correction applied to any of the patient's eyes; c.

Abstract: A method for creating a personalized E-Mail within a mail client, includes the creation of a signature as a part of the E-mail based on the interaction of a user with the mail client, a portrait image of the user being automatically integrated into the signature.

Abstract: A remotely controlled packable robot features a chassis with a top surface and a bottom surface, a pair of main tracks for maneuvering the chassis, and an open channel under the robot defined by the bottom surface of the chassis and the main tracks. A robot arm is foldable from a stored position in the open channel underneath the robot chassis to a deployed position extending upwards from the top surface of the chassis. A camera assembly may be foldable from a stowed position in the open channel underneath the robot chassis next to the robot arm to a deployed position extending upwards from the top surface of the chassis.

Abstract: A method includes generating a three-dimensional (3D) virtual reality video by stitching together image frames of an environment captured by a camera array. The method further includes generating graphical data for displaying a virtual reality user interface that includes the 3D virtual reality video. The method further includes determining, based on movement of a peripheral device, that a user moves a hand to be located in front of the 3D virtual reality video in the user interface and grabs the 3D virtual reality video from a first location to inside an object. The method further includes displaying the object with a preview of the 3D virtual reality video inside the object.

Abstract: A wearable electronic device displays an impact location that shows where a projectile fired from a weapon will hit a target and displays a bullseye location that shows a desired location where to hit the target. The wearable electronic device indicates firing the weapon when the impact location overlaps with the bullseye location.

Abstract: A wearable electronic device displays an impact location that shows where a projectile fired from a weapon will hit a target and displays a bullseye location that shows a desired location where to hit the target. The wearable electronic device indicates firing the weapon when the impact location overlaps with the bullseye location.

Abstract: An asymmetric video conferencing system comprises a panoramic video conferencing device in a first location and a head mounted video conferencing device in a second location. The panoramic video conferencing device acquires panoramic images of multiple participants via a 360 degree camera, identifies a face or other region of interest (ROI) in the panoramic images, and captures video data of the ROI. To reduce bandwidth utilized, the ROI video data is transmitted to the head mounted video conferencing device used by a remote participant, who can command alternative views and audio from the first location. Images and audio of the remote participant can be transmitted to the first location for viewing and hearing by the multiple participants, enabling virtual face-to-face discussions. A method implementing the system is also disclosed herein.

Abstract: A wireless vision equipment for weapons including an image capturing device configured to be mounted on a weapon using standard weapon attachment systems, a camera slot, an integrated transmitter using at least one wireless communication protocol, a beam splitter mirror aligned with an aim line, wherein the splitter mirror multiplies an incoming image by two conjugate images, a camera, a transmitter using the at least one wireless communication protocols, a receiver, an antenna, a cable, a shooting glass and a viewing device configured to be lifted upwards to avoid limiting a user's field of vision.

Abstract: Methods and apparatus for focusing in virtual reality (VR) or augmented reality (AR) devices based on gaze tracking information are described. Embodiments of a VR/AR head-mounted display (HMD) may include a gaze tracking system for detecting position and movement of the user's eyes. For AR applications, gaze tracking information may be used to direct external cameras to focus in the direction of the user's gaze so that the cameras focus on objects at which the user is looking. For AR or VR applications, the gaze tracking information may be used to adjust the focus of the eye lenses so that the virtual content that the user is currently looking at on the display has the proper vergence to match the convergence of the user's eyes.

Abstract: Humans may exhibit characteristic patterns of eye movements when looking at specific objects. For example, when a person looks at the face of another person, their eyes exhibit a certain pattern of movements and saccades as they look at the face. An electronic contact lens includes eye tracking sensors and an outward looking imaging system that may capture images of the user's environment. When the eye tracking sensors detect the pattern of eye movements characteristic of looking at a face, the imaging system becomes active and captures images and performs facial recognition to identify the face using the captured images. The results of the facial recognition may be displayed to the user using a projector of the electronic contact lens.

Abstract: A device may receive, from a plurality of sensors, sensor data relating to a user. The device may include a plurality of types of sensors including a spectrometer and one or more of an accelerometer, a heart rate sensor, a blood pressure sensor, a blood sugar sensor, a perspiration sensor, a skin conductivity sensor, or an imaging sensor. The device may process the sensor data, from the plurality of types of sensors, relating to the user to determine a health condition of the user. The device may provide, via a user interface, information identifying the health condition of the user based on processing the sensor data, from the plurality of types of sensors, relating to the user.

Abstract: A peripheral device that is used with a mixed reality system is thrown by the user of the mixed reality system and captures scene and mapping data of the area the peripheral is thrown into. Upon capturing the scene and mapping data, the peripheral transmits the data (in some embodiments, including processed results of evaluation of the data) to the mixed reality system. The mixed reality system displays a visual representation of the scene and mapping data on a near eye display. In some embodiments, the visual representation appears in real space enabling an effect where the user of the mixed reality system can see through walls or other obstructions. In some embodiments, the peripheral is configured to detonate upon the satisfaction of a variable set of conditions.

Abstract: Provided is a security management device for tracking a terminal located indoors. The security management device display on a map a position of the at least one wireless network camera and information about a terminal located in a range capable of wireless communication with the at least one wireless network camera, control a wireless network camera to track a position of the terminal based on the information about the terminal while the terminal is in a range capable of wireless communication with the wireless network camera, and control the wireless network camera to send identification information of the terminal to a neighboring wireless network camera when the terminal moves out of the range capable of wireless communication with the wireless network camera.

Abstract: A wearable proximity warning device is provided that uses a novel method of processing images from a high frame rate digital camera to detect human threats from behind and determine if there are any approaching threats by using novel pixel counting and threat detection analysis algorithms. The device is worn on the back of the body either by use of a belt clip or with chest straps. The user may use select from a variety of warning options from the device including audible warning tones, device vibration or smartphone SMS/MMS text messaging. Stored video is saved by the device and may be periodically uploaded to secure cloud storage. The device contains a rechargeable battery that may be recharged using a USB port. The device uses adaptive human recognition that switches between facial recognition detection mode and body recognition detection mode depending on distance of the threat from the user.

Abstract: A wearable proximity warning device called SIMS (Surrounding Intelligent Motion Sensor) is provided that uses a novel method of processing images from a high frame rate digital camera to detect human threats from behind and determine if there are any approaching threats by using novel pixel counting and threat detection analysis algorithms. The device is worn on the back of the body either by use of a belt clip or with chest straps. The user may use select from a variety of warning options from the device including audible warning tones, device vibration or smartphone SMS/MMS text messaging. Stored video is saved by the SIMS device and may be periodically uploaded to secure cloud storage. The device contains a rechargeable battery that may be recharged using a USB port.

Abstract: In one embodiment, a method includes receiving audio input during an audio-video communication session. The audio input is generated by a first sound source within an environment and a second sound source within the environment. The method includes receiving video input depicting the first sound source and the second sound source in the environment. The method includes identifying the first sound source and the second sound source using the audio input and the video input. The method includes predicting a first engagement metric for the first sound source and a second engagement metric for the second sound source based on the identifying. The method includes processing the audio input to generate an audio output signal based on a comparison of the first engagement metric and the second engagement metric. The method includes providing the audio output signal to a computing device associated with the audio-video communication session.