Abstract: Disclosed are various embodiments for processing verbal queries relative to video content. A verbal query that is associated with a portion of video content is received. The verbal query specifies a relative frame location. An action is performed based at least in part on the portion of the video content at the relative frame location.

Abstract: An environmental control system is provided. The environmental control system can include a cooling device and a heating device. The cooling device can lower the temperature of fluid contained within a cold reservoir and the heating device can raise the temperature of fluid contained within a hot reservoir.

Abstract: Systems and methods for efficient and secure management of encrypted “snapshots” for a remote provider substrate extension (“PSE”) of a cloud provider network substrate are provided. The PSE may request and obtain a snapshot from the cloud provider network substrate, restore a volume from the snapshot, make changes to data in the restored volume, and/or initiate the creation and storage of a new snapshot that includes incremental updates to the original snapshot to reflect the changes made to data in the volume. An encrypted snapshot stored within the cloud provider network substrate may be decrypted using a cloud provider key designed for internal use only, and then re-encrypted using a PSE-specific key before providing the snapshot to the PSE, thereby avoiding the sharing of the cloud provider internal use only key outside the cloud provider network substrate.

Abstract: Systems, methods, and computer-readable media are disclosed for container transporters and related item manipulation devices. In one embodiment, an example item manipulation device may include a support platform, and a conveyance surface attached to the support platform and configured to convey a container at least a portion of a distance from a first side of the support platform to a second side of the support platform, where the conveyance surface defines at least a portion of a container passage through which a container can be conveyed. The item manipulation device may include a manipulator comprising a first arm disposed along a longitudinal axis of the item manipulation device and a second arm disposed along the longitudinal axis of the item manipulation device, where the manipulator is engageable with the container to move the container at least partially onto the conveyance surface.

Abstract: Systems, methods, and computer-readable media are disclosed for dynamic adjustments to battery parameters using battery metrics. The device may be configured to determine a first value indicative of a battery voltage output during a first time interval, determine a second value indicative of a temperature during the first time interval, and determine a first acceleration factor for the battery during the first time interval based at least in part on the first value and the second value. The device may determine an adjusted number of charge cycles completed during the first time interval using the first acceleration factor, determine a total adjusted number of charge cycles of the battery, determine that the total adjusted number of charge cycles is equal to or greater than a first threshold, and cause the first maximum output voltage value to be reduced.

Abstract: Users can switch between applications using contextual interface elements. These elements can include icons for applications determined to likely be accessed by the user for a current context. Information is gathered to determine the current context, then information such as patterns of historical usage are utilized to determine and rank the applications by likelihood of use. Different contexts can include different icons, and a given context can include different icons for different points in time or locations. A user can access a contextual interface element by performing a swipe motion, for example. The user can continue the motion to an area associated with an icon of interest, and perform an action such as a tap or release to cause the associated application to be launched. Such an approach enables a user to quickly and easily launch another application independent of the application currently active on the device.

Abstract: A control plane server of a packet processing service assigns a first node group comprising exception-path nodes of the service to a network interface of a first application. Nodes of the assigned node group provide packet rewriting rules used by fast-path nodes of the service to direct requests of the application. In response to detecting that a workload migration criterion has been met, the control plane server initiates a migration workflow of the interface. During the workflow, flow state information of a packet flow is replicated at nodes of the first node group, sent from the first node group to a second node group via a fast-path node, and replicated at nodes of the second node group. Connections used for the application requests remain operational. After the migration workflow completes, nodes of the second node group provide packet rewriting rules for directing requests of the application.

Abstract: Technologies directed to arranging antenna elements in a triangular pattern on an antenna module of a phased array antenna are described. The phased array antenna includes a support structure and a first antenna module coupled to the support structure. The first antenna module element has a rectangular shape and includes a first set of antenna elements arranged as a first row and a second row within the rectangular shape. An antenna element of the first row and two antenna elements of the second row form a triangular pattern. Two adjacent antenna elements of the first set of antenna elements are separated by a first distance. Each antenna element of the first set of antenna elements has a first size that is less than half of the first distance.

Abstract: A probabilistic data structure such as a bloom filter may be used to provide a space efficient representation of rules that define events of interest to sources and may be used to reduce the number of events ingested by an event bus system. An event bus system may generate a probabilistic data structure based on the rules that may be used at the event source as a filter. The probabilistic data structure predetermines whether the events to be sent to the event bus system will be of interest to any event targets and if an event is of no interest to any of the event targets, the event will not be sent.

Abstract: A system and method for controlling a cooling system for an electronic datacenter component using a two-phase thermal management system with dynamic thermoelectric regulation. The system includes a thermoelectric cooler to transfer heat to a hot conduit of the thermal management system and initialize or maintain a natural convective flow of working fluid by maintaining a temperature difference between a hot and cold conduit.

Abstract: Systems and methods automatically generate a low input visualization of data updated in a workbook. One aspect describes a computer-implemented method comprising obtaining a request to update data in an application workbook for an application. The request comprises an indication of a type of update being requested and at least a subset of the data being updated. The method further comprises performing the update of the data in the application workbook and determining that the performed update of the data in the application workbook results in a change to a visualization of the data. The method also comprises generating an updated visualization based on the received request and the updated application workbook and generating a confirmation message that the updated visualization has been generated, wherein the updated visualization is generated automatically based on the update of the data.

Abstract: Disclosed are various embodiments for tracking developer behavior with respect to software analysis tools. In one embodiment, a security analysis is performed upon a first revision of a program, where the security analysis is based at least in part on a plurality of rules. A first security issue found in the security analysis upon the first revision of the program is identified. The security analysis is performed upon a second revision of the program. A second security issue found in the security analysis upon the second revision of the program is identified. The rules are updated based at least in part on whether the first security issue is corrected in the second revision as determined based at least in part on a comparison of the first security issue to the second security issue.

Abstract: Techniques for generating an item transmission portal are described herein. First information for a plurality of carriers may be maintained which includes a capacity for each carrier. Historical capacity requirements for geographic locations may be maintained. For each geographic location a forecast of future capacity requirements, for each week of a plurality of weeks, may be generated based on the historical capacity requirements. A plurality of documents may be generated for each geographic location based on the forecast and the first information. A value may be determined for each document based on a time and capacity commitment for each document, the historical capacity requirements, and the forecast. A user interface may be generated that includes the plurality of documents and corresponding determined values. The user interface may be updated to reflect acceptance of a specific document by a carrier in response to input received via the user interface.

Abstract: After a service receives a request from another service, the service determines an amount of time to process the request by the service as well as a remaining time allotment to complete processing the request (e.g., a timeout value). Based on the remaining time allotment and the amount of time to process the request by at least the service (predicted time or actual time), the service may determine whether to continue processing the request (e.g., by the service and/or one or more subsequent services) or fail the request. In response, the service may then continue processing the request (e.g., continue processing at the service itself or propagate the request to the next service), or the service may fail the request.

Abstract: Described are systems and methods for determining an agent that performed an event within a materials handling facility. A series of overhead images that include representations of the event location and one or more agents are processed to determine a motion or movement of the agent over a period of time. For example, a motion model representative of a motion of the agent over a period of time is generated from the images. A distance between the motion model and the event location is also determined. An association between the agent and the event may be determined based on the motion model and the distance between the motion model and the event location.

Abstract: Technologies directed to correction of terrestrial interference using hybrid beamforming are described. One method includes a first communication device with beamforming circuitry receiving an indication of a first direction towards a second communication device The method further includes determining that orienting a main lobe of an antenna gain pattern of the beamforming circuitry along the first direction results in RF saturation of the beamforming circuitry using first data that indicates a set of directions comprising the first direction and a status of an RF saturation condition corresponding to each of the set of directions. The method further includes using the first data to determine a second direction that is different from the first direction. Orienting the main lobe along the second direction does not result in RF saturation. The method further includes receiving, a first RF signal at a first time with the main lobe oriented in the second direction.

Abstract: Example embodiments provide techniques for configuring a system to receive an input at a source device and present an output at another device within the same space as the source device. The output device may be selected based on various factors relating to the capabilities of the output device, the type of output responsive to the user input, whether the output device is active, whether the output device was recently used and others.

Abstract: Techniques for dynamic compute allocation in multiple-bitrate live video streaming are described. As one example, a computer-implemented method includes receiving a request to perform a real-time encode of a live video at a content delivery service, dynamically allocating, by the content delivery service, a set of compute resources between parallel encoders to respectively encode different renditions of the live video based at least in part on real-time usage data, performing parallel real-time encoding of each of the different renditions of the live video by the set of compute resources of the content delivery service, and transmitting at least one of the encoded different renditions of the live video from the content delivery service to a viewer device.

Abstract: Described is a system for counting stacked items using image analysis. In one implementation, an image of an inventory location with stacked items is obtained and processed to determine the number of items stacked at the inventory location. In some instances, the item closest to the camera that obtains the image may be the only item viewable in the image. Using image analysis, such as depth mapping or Histogram of Oriented Gradients (HOG) algorithms, the distance of the item from the camera and the shelf of the inventory location can be determined. Using this information, and known dimension information for the item, a count of the number of items stacked at an inventory location may be determined.

Abstract: A lossy counter counts distinct network data items. The lossy counter includes a count sketch bounded by a predetermined value to limit the number of distinct network data items included in the count sketch. The count sketch may include counts for a set of distinct network data items. The lossy counter has an associated time interval, and the first set of distinct network data items and the second set of distinct data items include timestamps that coincide with the time interval associated with the lossy counter.

Abstract: This disclosure describes a user-recognition system that may perform one or more verification methods upon identifying a previous image that matches a current image of a palm of a user. For instance, the user-recognition system may perform the verification method(s) as part of the recognition method (e.g., after recognizing a matching image), in response to an audit process, in response to a request to re-analyze the image data (e.g., because a user indicates that he or she was not associated with a particular purchase or shopping session), and/or the like.

Abstract: Embodiments herein describe a combined aerial and ground sortation system. That is, the system can include both an aerial sortation system and a ground sortation system that work together to sort items (e.g., packages) in a warehouse or building. In one embodiment, the combined aerial and ground sortation system includes a leading sorter that identifies which packages should be sorted using the aerial sortation system and which should be sorted by the ground sortation system. The aerial sortation system may use drones to fly the package to one of the containers and drop the package into the container while the ground sortation system may use drive units that move along a floor of the warehouse to deposit the packages into the container. In another embodiment, the aerial sortation system performs a first, primary sort of the items while the ground sortation system performs a secondary sort.

Abstract: A design platform facilitates development and management of interactions and events involving objects in a virtual environment, such as a gaming environment. Each object, or type of object, can have one or more tags applied. The tags are used to determine a rule to enforce when an event occurs with respect to an object, or the object interacts with another object in the environment. This can include, for example, determining an animation, sound, or action to trigger for the object. A set of rules can be defined, where each rule is associated with a set of tags. A custom expression can provide logic to be evaluated to determine which rule to apply for an interaction, where the script can specify certain tags, or groups or tags, that must be present, or not present, for the rule to apply. The tags for a given object can be updated dynamically during gameplay.

Abstract: Users of a graph data store can access an index object for a graph structure in order to obtain information from objects in the graph data structured index by the index object. A query can be received that is directed to an index object. Links to a subset of data objects in the data store included in the index object can be used to access the subset of data objects to perform the query. In some embodiments, data objects can be manually added or removed from the links included in the index object.

Abstract: An imaging device may capture images of a scene, where the scene includes retroreflective materials. Where visual images and depth images are captured from a scene, and the depth images have ratios of supersaturated pixels that are less than a predetermined threshold, a location map of the scene is generated or updated based on the depth images. Where the ratios are greater than the predetermined threshold, the location map of the scene is generated or updated based on the visual images. Additionally, where each of a plurality of imaging devices detect concentrations of supersaturated pixels beyond a predetermined threshold or limit within their respective fields of view, an actor present on the scene may be determined to be wearing retroreflective material, or otherwise designated as a source of the supersaturation, and tracked with the scene based on coverage areas that are determined to have excessive ratios of supersaturated pixels.

Abstract: A system is provided for determining when supplemental content unresponsive to a user input is to be presented during a user interaction. The system determines an action, performance of which, may trigger output of supplemental content. The system may determine when, during a user interaction corresponding to performance of the action, supplemental content may be presented. The system may use constraint data indicating a device type via which the content may be presented, and a time duration during which the supplemental content may be presented. If the constraint data is satisfied, the system may determine to present the supplemental content.

Abstract: A system may generate image data of users within a facility. Such users may be part of a group that is unknown to the system. The system can predict group data for two or more users based on a resemblance or the users being within a threshold distance of each other. However, if a confidence level associated with the predicted group data is below a threshold value, the group data is deemed unreliable and assistance from an associate is deemed necessary. A user interface that includes a portion of the image data, information about the predicted group data, and other interface elements is presented to the associate via a display. Based on the input data received from the associate, the group data can be confirmed or rejected. If the group data is confirmed, an association is made between the users and a group identifier.

Abstract: Techniques for a context preservation during server restart are described. A restart of a database server having at least one open communications session with a database client is initiated. The database server can be executed by a compute instance hosted by a computer system of a provider network. A restarted database server is launched. The restarted database server obtains session state data to resume the at least one open communications session with the database client, the session state data including one or more session variables. The restarted database server receives a transaction from the database client.

Abstract: Technology is described for a checkpoint service for code function checkpoints associated with instances of code functions executed by a computing hub in a local area network. Subsequent to the execution of a code function being suspended by the computing hub, the computing hub may generate a code function checkpoint describing the execution of the code function. The computing hub may persist the code function checkpoint to storage using a checkpoint service. The computing hub may create the code function checkpoint in response to monitoring metrics associated with the computing hub, such as processor overutilization. The computing hub may resume the execution of the code function at a later time using the code function checkpoint retrieved from the checkpoint service.

Abstract: Techniques for data lake-based text generation and data augmentation for machine learning training are described. A user-provided dataset including documents and corresponding label information can be automatically supplemented by creating additional high-quality document samples, with labels, via a large repository of documents in a data lake. Documents from the data lake may be identified as being semantically similar to the user-provided documents but different enough to allow a resulting model to learn from the variation in these documents. New documents can be generated from user-provided document samples or data lake sample documents by identifying and replacing slots within the samples and rewriting adjunct tokens.

Abstract: High-speed processing of packets to, and from, a virtualization environment can be provided while utilizing hardware-based segmentation offload and other such functionality. A hardware vendor such as a network interface card (NIC) manufacturer can enable the hardware to support open and proprietary stateless tunneling in conjunction with a protocol such as single root I/O virtualization (SR-IOV) in order to implement a virtualized overlay network. The hardware can utilize various rules, for example, that can be used by the NIC to perform certain actions, such as to encapsulate egress packets and decapsulate packets.

Abstract: A database system may use a machine learning model creation system to create a machine learning model from data stored in the database system responsive to a request from a client. The database system may obtain an executable version of the machine learning model, based on an uncompiled hardware agnostic version of the machine learning model, according to the hardware configuration of one or more computing resources selected by the database system to perform requests to the database system that invoke the machine learning model to generate predictions.

Abstract: Systems and methods are described for modifying input and output (I/O) to an object storage service by implementing any combination of and any number of owner-specified functions and native functions. A function can implement a data manipulation. The functions can be applied prior to implementing a request method (e.g., GET, PUT, LIST, etc.) specified within the I/O request, such that the data to which the method is applied may not match the object specified within the request. For example, a user may request to obtain a data set. The data set may be passed to a native function that filters sensitive data to the data set, the output of the native function may be passed to an owner-specified function that redacts data from the filtered data set, and the request method may then be applied to the output of the owner-specified function.

Abstract: Systems and methods permit customers of a service provider network to specify various constraints on a desired fleet of virtual machine instances without having to specify the hardware types of instances to be included in the fleet. Instead, the customer can specify per-instance hardware constraints (number of CPUs, amount of memory, etc.) and job constraints (e.g., deadline, budget, application type, etc.). A provisioning service accesses an internal database containing instance cost data, instance availability data, and mappings between application type and fleet configurations to propose a fleet of instances that complies with the customer-specified per-instance hardware and job constraints, thereby freeing the customer from having to be conversant in the particular instances offered by the service provider.

Abstract: Techniques are described for enabling users of a cloud provider network to request the conditional termination of computing resources based on a forecasted availability of replacement capacity at a specified time or range of time in the future. A cloud provider network provides an application programming interface that can be used to make such requests, where the computing resources are hosted by the cloud provider network as part of a capacity pool shared by tenants of the cloud provider network. This type of request can be generated, for example, by a user desiring to terminate the use of some number of unproductive computing resources only if the user can be reasonably assured by the cloud provider network that capacity will be available at a future time when the user will likely need the capacity again.

Abstract: Results of different programmatic interfaces may be paginated. A request that causes aggregation of results separately returned from different programmatic interfaces may be received. Attempts to obtain the results using the different programmatic interfaces to execute the request may be made. A token that describes result states for the different programmatic interfaces may be included in a response to the request.

Abstract: An industrial monitoring system comprises a monitoring device attached to an industrial device by a bond. Sensor data collected by the monitoring device during a commissioning period is received and used to train a machine learning model. Subsequent to the commissioning period, additional sensor data is collected by the monitoring device. An abnormal state of the bond between the monitoring device and industrial device is determined based on the additional sensor data and a characteristic inferred by the trained machine learning model. A notification of the abnormal state is generated.

Abstract: Techniques are provided herein for modifying content playback between user devices (e.g., moving the content playback from one device to another). State information may be obtained by a content management engine, the state information describing aspects of historical content streaming corresponding to a plurality of user devices associated with a user profile. A streaming request can be received from a user device. The content to be streamed can be selected by the content management engine based at least in part on the state information. In some embodiments, the playback device is also selected based on the state information. A streaming request is transmitted from the content management engine to a service provider to effectuate the content being streamed to the selected playback device. These techniques enable a user to push or pull content to or from one device to another.

Abstract: Identifying a person committing a crime in video data captured by a security device. An information request message identifying the video data and a need for information about the video data is received. When the video data is determined suitable for sharing within a geographic network, an access control of the video data is set to allow a client device registered with the geographic network to display the video data. A display control value of the video data is set to direct display of a label with the video data to indicate the need for the information. A rating of usefulness of the information received from the client device is determined and a first value is added to an account associated with the client device based at least in part upon the rating.

Abstract: Devices, systems, and methods are provided for goal-based audience selection. A method for generating an audience using machine learning may include receiving a request to generate an audience for an advertisement campaign, the request including an objective associated with presentation of the advertisement campaign. The method may include determining first user actions based on the objective, and identifying first users of a system who performed the first user actions using the system. The method may include determining second user actions performed by the first users prior to performing the first user actions, and identifying second users of the system who performed the second user actions and failed to perform the first user actions. The method may include generating a target audience to which to present the advertisement campaign, and causing presentation of the advertisement campaign to the target audience.

Abstract: Techniques for storing encrypted data using a storage service system are described herein. A computer system of a computation layer of the storage service system receives an encrypted key manifest, which is then decrypted using a cryptoprocessor of the computer system of the computation layer to produce a partition key. The partition key is then provided to a file system abstraction layer so that, as data is provided to the computation layer for storage, the file system abstraction layer can use the partition key to encrypt data and store the encrypted data in the storage layer.

Abstract: Techniques are described for automating virtual placements in video content.

Abstract: A speech-capture device can capture audio data during wakeword monitoring and use the audio data to determine if a user is present nearby the device, even if no wakeword is spoken. Audio such as speech, human originating sounds (e.g., coughing, sneezing), or other human related noises (e.g., footsteps, doors closing) can be used to detect audio. Audio frames are individually scored as to whether a human presence is detected in the particular audio frames. The scores are then smoothed relative to nearby frames to create a decision for a particular frame. Presence information can then be sent according to a periodic schedule to a remote device to create a presence “heartbeat” that regularly identifies whether a user is detected proximate to a speech-capture device.

Abstract: Techniques for controlling a robotic picking arm using estimated seal quality metrics. A plurality of candidate contact points for holding an item using a suction device of the robotic picking arm, based on captured images of the item and an n-dimensional surface model of the item. An expected seal quality metric for a first one of the candidate contact points, by processing the n-dimensional surface model of the item and physical properties of the suction device of the robotic picking arm. Based on the expected seal quality metric, embodiments can determine whether to retrieve the item from the storage container by holding the item at the first candidate contact point using the suction device of the robotic picking arm.