Abstract: The present disclosure is drawn to devices and systems for target detection in samples, particularly allergen detection in food samples. The allergen detection system includes a sampler, a disposable analysis cartridge and a detection device with an optimized optical system. The allergen detection utilizes nucleic acid molecules as detection agents and detection probes.

Abstract: Disclosed is a magnification device with multiple orientations. This allows a blind or low vision user to select the best orientation for the task being performed. The device includes both opened and closed orientations. It is further configurable into document reading, near object inspection, far object inspection, and self-inspection modes. The camera of the device can be manipulated by the user depending upon which of these orientations is selected. Furthermore, the camera automatically adjusts itself into one of two pre-set configurations whenever the device is opened or closed.

Abstract: In some embodiments, a method of combining two or more songs to present one or more campaigns to a plurality of audience members is provided. The method can include combining a first music selection, a second music selection, and one or more videos into a campaign video. The method can further comprise providing the campaign video for playback to the plurality of audience members and receiving a vote from the plurality of audience members, the vote reflecting a preference for the first selection or the second selection.

Abstract: An unmanned aerial vehicle (UAV) network cell that uses modular communication modems may be configured to support various communication standards and communication frequency bands. A UAV communication controller may monitor a signal robustness value for a communication frequency band that a UAV network cell is using for a relay backhaul with a ground network cell of a wireless carrier network. In response to determining that the signal robustness value of the communication frequency band has dropped below a predetermined threshold, the UAV communication controller may command the UAV network cell to use an additional communication frequency band that is different from the communication frequency band to carry at least one portion of backhaul communication with the wireless carrier network.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for performing action localization in images and videos. In one aspect, a system comprises a data processing apparatus; a memory in data communication with the data processing apparatus and storing instructions that cause the data processing apparatus to perform image processing and video processing operations comprising: receiving an input comprising an image depicting a person; identifying a plurality of context positions from the image; determining respective feature representations of each of the context positions; providing a feature representation of the person and the feature representations of each of the context positions to a context neural network to obtain relational features, wherein the relational features represent relationships between the person and the context positions; and determining an action performed by the person using the feature representation of the person and the relational features.

Abstract: A method for determining an architecture for a task neural network configured to perform a particular machine learning task is described.

Abstract: An unmanned aerial vehicle (UAV) network cell that uses modular communication modems may be configured to support various communication standards and communication frequency bands. An authentication request from a modular communication modem of a plurality of modular communication modems in an unmanned aerial vehicle (UAV) network cell may be received at a core network of a wireless carrier network following an installation of the modular communication modem into the UAV network cell, in response to authenticating that the modular communication modem is authorized to communicate with the core network, the modular communication modem is registered at the core network to handle communication between the UAV network cell and the core network. However, in response to determining that the modular communication modem is not authorized to communicate with the core network, the modular communication modem is disallowed from handling communication between the UAV network cell and the core network.

Abstract: The present invention is drawn to devices and systems for allergen detection in food samples. The allergen detection system includes a disposable analysis cartridge and a detection device with an optimized optical system.

Abstract: A bassinet includes a base, a column assembly rotatably connected to the base at a first end of the column assembly, an arm rotatably connected to a second end of the column assembly, a tub frame pivotably connected to the arm, an infant tub retained within the tub frame, the infant tub including a front wall, a back wall, and two side walls and having areas of reduced height along the front wall and at least one of the side walls to provide enhanced access to the infants located within the infant tub, and an insert configured to reversibly fit within an interior of the infant tub.

Abstract: A wall mounting bracket assembly can be adapted to support an electronic display. The bracket assembly can include a support frame that includes a first support rail, and the support frame can be adapted to couple to a vertical surface. The bracket assembly can include one or more leveling assemblies. The one or more leveling assemblies can include a hook portion adapted to engage with the first support rail. The one or more leveling assemblies can include a first pair of support plates and a first leveling plate positioned between the first pair of support plates. The first pair of support plates can be adapted to translate with respect to the first leveling plate. The one or more leveling assemblies can include a driving block configured to engage with the first leveling plate and to translate the first pair of support plates with respect to the first leveling plate.

Abstract: Disclosed is a tablet with an improved Braille display. The Braille display employs a pin array that allows for the selective use of either six or eight pin cells. This is accomplished by turning off or on a pin pair adjacent to each cell. The spacing of the pins also allows capacitive sensors to be located adjacent to each Braille cell. These sensors are used to determine the location of the user's finger upon the display. The pin spacing further allows geometric shapes to be generated in additional to text.

Abstract: A method for determining an architecture for a task neural network configured to perform a particular machine learning task is described.

Abstract: Systems and methods for transforming grayscale images into color images using deep neural networks are described. One of the systems include one or more computers and one or more storage devices storing instructions that, when executed by one or more computers, cause the one or more computers to implement a coloring neural network, a refinement neural network, and a subsystem. The coloring neural network is configured to receive a first grayscale image having a first resolution and to process the first grayscale image to generate a first color image having a second resolution lower than the first resolution. The subsystem processes the first color image to generate a set of intermediate image outputs. The refinement neural network is configured to receive the set intermediate image outputs, and to process the set of intermediate image outputs to generate a second color image having a third resolution higher than the second resolution.

Abstract: A testing platform tests an electrical and mechanical system such as an HVAC unit according to an algorithm that reduces the total testing time of the components of the system, while ensuring the safety of the system during system-wide testing. The platform uses constraints that are checked both before and during the testing to ensure that HVAC operating conditions are acceptable for starting and maintaining component tests. Preferably, the platform uses finite-state machines for each device to organize the component tests, allowing for monitoring of constraints and starting, pausing, and stopping component tests. Preferably, total test execution time is reduced by running component tests in parallel, running component tests based on loads of the components, or combinations of both.

Abstract: Systems and methods to manage incentive campaigns and automatically approve requests for incentives received from consumers or their proxy based on dynamic parameters defined by an incentive provider are described herein. Users at various levels between an original equipment manufacturer and the consumer may interact and offer input. In various implementations, campaign parameters may be received from an incentive provider and used to initiate and/or otherwise administrate a campaign. Incentive providers may then dynamically modify ongoing campaigns via the system. One or more user interfaces may be generated through which users may interact with the system and visualizations related to a campaign may be presented. Based on the initial or modified parameters for a campaign, the systems and methods described herein may further determine whether to approve, deny, or otherwise respond to requests for incentives received from consumers or their proxy.

Abstract: A method for generating stereoscopic imagery comprising the steps of building an automated benthic ecology system comprising a high-resolution still camera, a high-resolution video camera, an environmental sensor package, a stereoscopic camera, and underwater housing, wherein the high-resolution video camera is configured for constant recording, turning the video camera on and hovering the ABES at the desired depth for a specific period of time acquiring imagery, wherein the imagery is automatically time-stamped with a date and time, extracting frames from the video, matching the timestamps on the photographs with the timestamps of the ROV log, which provides latitude, longitude and depth measurements, georeferencing each image using GeoSetter software, stitching imagery from the stereoscopic camera into a panoramic image using software.

Abstract: Disclosed is a portable magnifier camera with both storage and deployed configurations. In the storage configuration, the entire assembly can be carried via an associated handle. The camera is designed to rest on a desktop, or other planar surface, in the deployed configuration. When deployed, a housing and camera arm are positioned in a cantilevered arrangement over top of an object to be viewed. A camera housing, with an associated camera, is pivotally connected to the camera arm and is positioned over the object upon the camera being placed into a deployed configuration.

Abstract: Disclosed is a braille display that is impact resistant and that employs a modular construction. The impact resistance is achieved, in part, by molded and resilient end caps. Each end cap includes a centrally positioned slit that functions in absorbing lateral impacts to the display. The impact resistance is further achieved by an over molded housing that is positioned about an associated USB connector. The modularity of the display is achieved by way of a subassembly that is releasably secured within the interior of a main housing. The subassembly, in turn, includes a series of cell compartments that are arranged in groups. Each group of cell compartments is controlled by an individual shift register. This allows a grouping cell compartments to be removed and replaced as needed without disturbing the remaining cell components.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for implementing a machine-learned object tracking policy. One of the methods includes receiving a current video frame by a user device having a plurality of installed object trackers, wherein each object tracker is configured to perform a different object tracking procedure on the current video frame rent video frame. The current video frame and one or more object tracks previously generated by the one or more object trackers are provided as input to a trained policy engine that implements a reinforcement learning model to generate a particular object tracking plan. A particular object tracking plan is selected based on the output of the reinforcement learning model, and the selected object tracking plan is performed on the current video frame to generate one or more updated object tracks for the current video frame.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for performing action localization. In one aspect, a system comprises a data processing apparatus; a memory in data communication with the data processing apparatus and storing instructions that cause the data processing apparatus to perform operations comprising: receiving an input comprising an image depicting a person; identifying a plurality of context positions from the image; determining respective feature representations of each of the context positions; providing a feature representation of the person and the feature representations of each of the context positions to a context neural network to obtain relational features, wherein the relational features represent relationships between the person and the context positions; and determining an action performed by the person using the feature representation of the person and the relational features.