Abstract: Aspects of the disclosed technology provide solutions for searching objects within multimedia content based on multi-modal embeddings. An example method can include receiving media content including a plurality of video frames. The method can include steps for generating, using a pre-output layer of a machine learning algorithm, one or more multimodal feature embeddings describing at least one object for the plurality of video frames, receiving a query including a request to search the media content for a matching object, determining whether the media content includes the matching object based on the one or more multimodal feature embeddings describing the at least one object, and returning one or more results in response to determining that the media content includes the matching object. Systems and machine-readable media are also provided.

Abstract: Aspects of the disclosed technology provide solutions for customizing audio streams for individual users. An example process can include steps for receiving an audio stream, establishing a connection with a first audio device associated with a first user, establishing a connection with a second audio device associated with a second user, and delivering a first segment of the audio stream to the first audio device based on user preferences associated with the first user. The process can further include steps for delivering a second segment of the audio stream to the second audio device based on user preferences associated with the second user. Systems and machine-readable media are also provided.

Abstract: System, apparatus, article of manufacture, method and/or computer program embodiments are provided for determining a presentation configuration of content. An example method can include determining information about primary content displayed on a first display and an environment of the first display, the information including a display configuration of the primary content, a characteristic of the primary content, and/or an indication whether a user is present in the environment and/or the environment includes a second display coupled to a client or a different client that is coupled to the client; based on the information, determining a different display configuration for displaying the secondary content; and based on the different display configuration, generating an instruction to display the secondary content at the second display or on an ROI at the first display, the ROI excluding the primary content or including a portion of the primary content having a saliency below a threshold.

Abstract: Aspects of the disclosed technology provide solutions for controlling multiple displays (e.g., multiple display devices, multiple displays on a media device, etc.) using a remote control device. An example method can include assigning a code for each of a plurality of display devices based on an image of the plurality of display devices. In some instances, the code of each of the plurality of display devices enables a remote control to independently operate each of the plurality of display devices. Moreover, the example method can include determining a position of the remote control relative to each of the plurality of display devices based on the image. Further, the example method can include determining a display device of the plurality of display devices for the remote control to control based on the position of the remote control relative to each of the plurality of display devices.

Abstract: System and techniques are provided for calibrating lighting systems. An example method can include determining, based on data from a sensor in a scene, different light outputs generated by a light source in the scene, the different light outputs including brightness, intensity, and/or color levels; determining a curve plotting a relationship between the different light outputs and different dimming inputs used to control the light source; based on the curve, mapping the different dimming inputs to control signals configured to trigger the light source to generate the different light outputs, the mapping correlating each dimming input to a light output and a control signal configured to trigger the light source to generate the light output; and based on the mapping, calibrating each dimming input to control the light source to generate the light output associated with that dimming input based on the control signal correlated to that dimming input.

Abstract: Aspects of the disclosed technology provide solutions for dynamically automating a security system using machine learning. An example method can include receiving sensor data collected by a sensor installed outside of an indoor location. The sensor data may include an indication of a motion event occurring within a predetermined distance from the indoor location. The method can include, based on user data associated with the indoor location, predicting, using a neural network, a user behavior in response to the motion event. The method can further include, based on the predicted user behavior, determining, using the neural network, an action comprising a response to the motion event implemented by one or more devices and automatically activating at least one of the device(s) to perform the action.

Abstract: Aspects of the disclosed technology provide solutions for searching objects within multimedia content based on multi-modal embeddings. An example method can include receiving media content including a plurality of video frames. The method can include steps for generating, using a pre-output layer of a machine learning algorithm, one or more multimodal feature embeddings describing at least one object for the plurality of video frames, receiving a query including a request to search the media content for a matching object, determining whether the media content includes the matching object based on the one or more multimodal feature embeddings describing the at least one object, and returning one or more results in response to determining that the media content includes the matching object. Systems and machine-readable media are also provided.

Abstract: Systems, methods, and computer-readable media are provided for content item positioning. In some examples, a method can include obtaining a first content item for display at a first display device, the first content item comprising video data; based on the video data, generating a saliency map of the first content item, the saliency map identifying regions of the first content item, each region being associated with a saliency value; determining, based on the saliency map, whether one or more regions of the regions have a saliency value that is below a predetermined saliency value; and based on the determining whether the one or more regions have a saliency value that is below a predetermined saliency value, determining whether to insert a second content item within the one or more regions of the first content item or within a display region of a second display device.

Abstract: Aspects of the disclosed technology provide solutions for controlling multiple displays (e.g., multiple display devices, multiple displays on a media device, etc.) using a remote control device. An example method can include receiving an image of a plurality of display devices. The method can include steps for analyzing the image to determine locations of the plurality of display devices, assigning a code specific to each of the plurality of display devices based on the image, and transmitting the code to a remote control. The code enables the remote control to independently operate each of the plurality of display devices.

Abstract: Disclosed herein are computing system, apparatus, article of manufacture, method and/or computer program product embodiments, and/or combinations and sub-combinations thereof, for distributing processing of sensor data. For example, a computing system may be configured to determine processing capabilities of each of one or more edge devices of a network. Additionally, the computing system may determine available processing resources of each of the one or more edge devices. Moreover, the computing system may select, from the one or more edge devices, a target device, based on the processing capabilities of each of the one or more edge devices and the available processing resources of each of the one or more edge devices. Further, the computing system may communicate with the target device to cause the target device to generate one or more feature vectors based on sensor data generated by the target device.

Abstract: Disclosed herein are system, apparatus, article of manufacture, method and/or computer program product embodiments, and/or combinations and sub-combinations thereof, for performing event or object classification. An example process can include receiving a first trigger corresponding to a first motion event within a field of view of a first image sensor; selecting a first video frame from a sequence of video frames captured by the first image sensor, wherein the first video frame is captured prior to the first trigger; selecting a second video frame from the sequence of video frames, wherein the second video frame is captured after the first trigger; determining at least one difference between the first video frame and the second video frame; determining, based on the at least one difference, at least one of an object classification and an event classification; and generating a notification that corresponds to the object classification or the event classification.

Abstract: Disclosed herein are system, apparatus, article of manufacture, method and/or computer program product embodiments, and/or combinations and sub-combinations thereof, for altering the appearance of electronic devices. An example process can include obtaining a visibility configuration for altering an appearance of an exterior surface of an electronic device; and modifying at least a portion of the exterior surface of the electronic device based on the visibility configuration.

Abstract: Aspects of the disclosed technology provide solutions for controlling multiple displays (e.g., multiple display devices, multiple displays on a media device, etc.) using a remote-control device. An example method can include receiving, at a remote-control device, a first wireless signal from a first multimedia device and determining a first signal strength of the first wireless signal. The method includes receiving, at the remote-control device, a second wireless signal from a second multimedia device and determining a second signal strength of the second wireless signal. The method also can include identifying a target multimedia device among the first multimedia device and the second multimedia device based on the first signal strength and the second signal strength and selecting a wireless communication format for communication between the remote-control device and the target multimedia device.

Abstract: Aspects of the disclosed technology provide solutions for controlling multiple displays (e.g., multiple display devices, multiple displays on a media device, etc.) using a remote control device. An example method can include receiving an image of a plurality of display devices. The method can include steps for analyzing the image to determine locations of the plurality of display devices, assigning a code specific to each of the plurality of display devices based on the image, and transmitting the code to a remote control. The code enables the remote control to independently operate each of the plurality of display devices.

Abstract: Aspects of the disclosed technology provide solutions for controlling multiple displays (e.g., multiple display devices, multiple displays on a media device, etc.) using a remote control device. An example method can include receiving an image of a plurality of display devices. The method can include steps for analyzing the image to determine locations of the plurality of display devices, assigning a code specific to each of the plurality of display devices based on the image, and transmitting the code to a remote control. The code enables the remote control to independently operate each of the plurality of display devices.

Abstract: System, apparatus, article of manufacture, method and/or computer program embodiments are provided for determining a presentation configuration of content. An example method can include determining information about primary content displayed on a first display and an environment of the first display, the information including a display configuration of the primary content, a characteristic of the primary content, and/or an indication whether a user is present in the environment and/or the environment includes a second display coupled to a client or a different client that is coupled to the client; based on the information, determining a different display configuration for displaying the secondary content; and based on the different display configuration, generating an instruction to display the secondary content at the second display or on an ROI at the first display, the ROI excluding the primary content or including a portion of the primary content having a saliency below a threshold.

Abstract: Disclosed herein are system, method and/or computer program product embodiments, and/or combinations and sub-combinations thereof, for generating a long exposure composite image (LECI) frame, wherein the LECI frame provides a summary of activity in a monitored environment over a duration of time. An embodiment captures, using an image-capturing device, a sequence of image frames over the duration of time using an optical sensor of the image-capturing device. The embodiment then generates the LECI frame, using an embedded low-power processor coupled to the optical sensor and dedicated to generating LECI frames, by combining the sequence of image frames. The embodiment then transmits the LECI frame to a network device using a transceiver coupled to the low-power processor.

Abstract: Disclosed herein are system, apparatus, article of manufacture, method and/or computer program product embodiments, and/or combinations and sub-combinations thereof, for implementing context-aware device configuration profiles. An example method can include determining a context associated with a client device; based on the context, determining to reduce a use of one or more resources of the client device by one or more components of the client device while the client device operates in the context, the one or more resources including a memory, one or more processors, one or more processor cores, and/or a power source; and in response to determining to reduce the use of the one or more resources by the one or more components while the client device operates in the context, adjusting a configuration profile of the client device, the adjusted configuration profile reducing the use of the one or more resources by the one or more components.

Abstract: Disclosed herein are computing system, apparatus, article of manufacture, method and/or computer program product embodiments, and/or combinations and sub-combinations thereof, for distributing processing of sensor data. For example, a computing system may be configured to determine processing capabilities of each of one or more edge devices of a network. Additionally, the computing system may determine available processing resources of each of the one or more edge devices. Moreover, the computing system may select, from the one or more edge devices, a target device, based on the processing capabilities of each of the one or more edge devices and the available processing resources of each of the one or more edge devices. Further, the computing system may communicate with the target device to cause the target device to generate one or more feature vectors based on sensor data generated by the target device.

Abstract: Disclosed herein are system, apparatus, article of manufacture, method and/or computer program product embodiments, and/or combinations and sub-combinations thereof, for processing image and video content. An example process can include obtaining a first image frame and a second image frame corresponding to a field of view of an image sensor; identifying at least one occlusion within the second image frame based on a comparison between the first image frame and the second image frame, wherein the at least one occlusion obfuscates a portion of the field of view of the image sensor; and removing the at least one occlusion from the second image frame to yield a revised second image frame, wherein the revised second image frame includes prior image data within an area corresponding to the portion of the field of view obfuscated by the at least one occlusion.