Abstract: Systems and methods are provided for training a depth model to recover scale factor for self-supervised depth estimation in monocular images. According to some embodiments, a method comprises receiving an image representing a scene of an environment; deriving a depth map for the image based on a depth model, the depth map comprising depth values for pixels of the image; estimating a first scale for the image based the depth values; receiving depth data captured by a range sensor, the depth data comprising a point cloud representing the scene of the environment, the point cloud comprising depth measures; determining a second scale for the point cloud based on the depth measures; determining a scale factor based the second scale and the first scale; and updating the depth model based on the scale factor, wherein the depth model generates metrically accurate depth estimates based on the scale factor.

Abstract: Text blocks are semantically compared, and a semantic score is provided to a user. The semantic score is based on application of a machine learning model trained on a text corpus. One or both of the two text blocks may have one or more words that do not appear in the training text corpus (skip-words). Skip-words are used, rather than discarded, to adjust the semantic score via, for example, a penalization function. The user provides feedback about the accuracy of the adjusted semantic score, and the feedback is used to perform supervised learning model.

Abstract: Aspects of the disclosure relate to performing enhanced exception processing using cognitive automation tools. In some embodiments, a computing platform may receive interaction data identifying one or more actions performed by one or more users in resolving a plurality of exception items associated with an exception queue. Subsequently, the computing platform may train, using a learning engine, a machine learning model to resolve a first exception and a second exception of one or more exceptions based on the interaction data. Based on training the machine learning model, the computing platform may generate one or more configuration commands directing a processing module to implement the machine learning model to process additional exception items associated with the exception queue. The computing platform then may send the one or more configuration commands to the processing module.

Abstract: A system is provided for file capture and in-line error correction. The system comprises: a controller configured to capture and process a file during a file capture sequence comprising one or more capture events, the controller being configured to: capture a file using a file capture device associated with a user device, wherein capturing the file comprises receiving a file image and a user input data field associated with the file; identify, using an in-line file analysis module, an error with a capture event during the file capture sequence, wherein the error is a discrepancy identified between the user input data field and an image-derived data field; correct the error in the file before the file capture sequence is completed and the file is exported; and export the file, wherein the file capture sequence is completed.

Abstract: A metal coordination polymer, in particular, a layered metal coordination polymer, can be used as a precursor to form nanostructures of various morphologies and composition. Metal based nanostructures can be prepared from the metal coordination polymers. The nanostructures may have various catalytic properties. The layered metal coordination polymer includes two or more layers, each layer including metal atoms coordinated to an organic linker to form a metal coordination polymer layer.

Abstract: Techniques for managing sequencing requests for storage node operations based on types of operations being sequenced. The techniques manage sequencing requests for different types of operations, such as backup and recovery operations, replication operations, migration operations, and so on, in a manner that avoids overwhelming storage node capacity. The techniques include receiving a sequencing request for an operation of a specified operation type at a request manager of a storage node and determining whether a capacity of the storage node is available for running the operation by a capacity manager for the specified operation type, in which the capacity manager is provided by a storage client and implemented on the storage node separate from the request manager. In this way, throttling, limiting, and/or prioritization requirements of the operation of the specified operation type can be defined by the storage client in a manner agnostic to the request manager.

Abstract: Animal litter boxes are provided which include a sensor to detect a change in an environment within an interior of its housing and an odor removal device (ORD) operationally connected to the sensor such that the ORD is activated in response to a signal from the sensor. In some cases, the ORD is configured to operate in different manners based on respectively different signals from the sensor. In addition or alternatively, the litter boxes may include a timer configured with the sensor for activating and periodically reactivating the ORD at preset time intervals in response to a sensor signal. In yet other embodiments, the litter boxes may additionally or alternatively include a timer having user adjustable settings of different time periods at which to initiate activation of the ORD and/or user adjustable settings of different durations at which to operate the ORD, both in response from a sensor signal.

Abstract: Aspects of the disclosure relate to performing enhanced deposit item processing using cognitive automation tools. In some embodiments, a computing platform may receive, from a deposit item input support server, image data associated with a deposit item. Subsequently, the computing platform may apply a machine learning classification model to the image data associated with the deposit item. In doing so, the computing platform may produce one or more predicted values for one or more fields of the deposit item and one or more confidence scores for the one or more predicted values. Based on these predicted values and confidence scores, the computing platform may generate and send one or more commands directing the deposit item input support server to accept or reject the deposit item, which in turn may cause the deposit item input support server to accept or reject the deposit item.

Abstract: Disclosed herein are systems and devices for performing computer-aided surgical navigation. Specifically, a system for augmented reality assisted trauma fixation, which displays various views of one or more tracked items, including a guide device, a fixation plate, a surgical tracking device, and an augmented reality display. The guide device having a tip, a handle, and a fixation guide. The fixation plate having at least one aperture designed to receive the tip of the guide. The surgical tracking device configured to track one or more trackable objects associated with the guide, fixation plate, and display, and thereby calculate a location of a fixation device, such as a screw, relative to the fixation plate.

Abstract: The disclosure relates to medical tubes and methods of manufacturing medical tubes. The tube may be a composite structure made of two or more distinct components that are spirally wound to form an elongate tube. For example, one of the components may be a spirally wound elongate hollow body, and the other component may be an elongate structural component also spirally wound between turns of the spirally wound hollow body The tube need not be made from distinct components, however. For instance, an elongate hollow body formed (e.g., extruded) from a single material may be spirally wound to form an elongate tube. The elongate hollow body itself may in transverse cross-section have a thin wall portion and a relatively thicker or more rigid reinforcement portion. The tubes can be incorporated into a variety of medical circuits or may be employed for other medical uses.

Abstract: Methods and apparatus for efficiently fulfilling product orders corresponding are described. Order items are divided into those which are to be automatically picked using an automated pick cell including a loader, such as a robotic arm, storage buffers and an apparatus for transferring the automatically picked items onto a robotic cart. The manually picked items are picked by a human picker guided by the robotic cart. The manual pick can occur before or after the automatic pick of items. In some embodiments whether items are to be automatically picked before or after the manual pick portion of an order depends on which ordering will allow for packing of the items corresponding to an order into the least number of receptacles, e.g., totes which are transported by the robotic pick carts.

Abstract: A system identifies a first text fragment as being under evaluation, wherein the first text is in at least a first document being compared to a second document using a first semantic model. The system compares the first text fragment to one or more text fragments of the second document. The system identifies top-k text fragments in the second document that are most similar to the first text fragment based on processing using the first semantic model. The system presents a user, via a graphical user interface (GUI), the top-k text fragments in visual proximity to the first text fragment.

Abstract: The disclosure relates to medical tubes and methods of manufacturing medical tubes. The tube may be a composite structure made of two or more distinct components that are spirally wound to form an elongate tube. For example, one of the components may be a spirally wound elongate hollow body, and the other component may be an elongate structural component also spirally wound between turns of the spirally wound hollow body The tube need not be made from distinct components, however. For instance, an elongate hollow body formed (e.g., extruded) from a single material may be spirally wound to form an elongate tube. The elongate hollow body itself may in transverse cross-section have a thin wall portion and a relatively thicker or more rigid reinforcement portion. The tubes can be incorporated into a variety of medical circuits or may be employed for other medical uses.

Abstract: A material handling system has a vertical lift two vertically aligned carriages that are independently positionable on a vertical support structure by respective drive systems. A processor subsystem selects one of the two carriages to move an article from at least one in-feed conveyor to a selected level of an automated storage and retrieval system (AS/RS). The processor subsystem concurrently positions the other carriage to avoid a collision.

Abstract: Some embodiments provide for an inspiratory limb for a breathing circuit that includes a first segment that comprises a first heater wire circuit and a second segment that comprises a second heater wire circuit. The inspiratory limb can include an intermediate connector that includes a connection circuit that electrically couples the first heater wire circuit to the second heater wire circuit. The inspiratory limb can be configured to operate in two modes wherein, in a first mode, electrical power passes through the first electrical connection to provide power to the first heater wire circuit without providing power to the second heater wire circuit, and in a second mode, electrical power pass through the first electrical connection to provide power to both the first heater wire circuit and the second heater wire circuit.

Abstract: A material handling system has a vertical lift two vertically aligned carriages that are independently positionable on a vertical support structure by respective drive systems. A processor subsystem selects one of the two carriages to move an article from at least one in-feed conveyor to a selected level of an automated storage and retrieval system (AS/RS). The processor subsystem concurrently positions the other carriage to avoid a collision.

Abstract: Aspects of the disclosure relate to performing enhanced exception processing using cognitive automation tools. In some embodiments, a computing platform may receive interaction data identifying one or more actions performed by one or more users in resolving a plurality of exception items associated with an exception queue. Subsequently, the computing platform may train, using a learning engine, a machine learning model to resolve a first exception and a second exception of one or more exceptions based on the interaction data. Based on training the machine learning model, the computing platform may generate one or more configuration commands directing a processing module to implement the machine learning model to process additional exception items associated with the exception queue. The computing platform then may send the one or more configuration commands to the processing module.

Abstract: The disclosure relates to medical tubes and methods of manufacturing medical tubes. The tube may be a composite structure made of two or more distinct components that are spirally wound to form an elongate tube. For example, one of the components may be a spirally wound elongate hollow body, and the other component may be an elongate structural component also spirally wound between turns of the spirally wound hollow body The tube need not be made from distinct components, however. For instance, an elongate hollow body formed (e.g., extruded) from a single material may be spirally wound to form an elongate tube. The elongate hollow body itself may in transverse cross-section have a thin wall portion and a relatively thicker or more rigid reinforcement portion. The tubes can be incorporated into a variety of medical circuits or may be employed for other medical uses.

Abstract: Aspects of the disclosure relate to performing enhanced deposit item processing using cognitive automation tools. In some embodiments, a computing platform may receive, from a deposit item input support server, image data associated with a deposit item. Subsequently, the computing platform may apply a machine learning classification model to the image data associated with the deposit item. In doing so, the computing platform may produce one or more predicted values for one or more fields of the deposit item and one or more confidence scores for the one or more predicted values. Based on these predicted values and confidence scores, the computing platform may generate and send one or more commands directing the deposit item input support server to accept or reject the deposit item, which in turn may cause the deposit item input support server to accept or reject the deposit item.

Abstract: Some embodiments provide for an inspiratory limb for a breathing circuit that includes a first segment that comprises a first heater wire circuit and a second segment that comprises a second heater wire circuit. The inspiratory limb can include an intermediate connector that includes a connection circuit that electrically couples the first heater wire circuit to the second heater wire circuit. The inspiratory limb can be configured to operate in two modes wherein, in a first mode, electrical power passes through the first electrical connection to provide power to the first heater wire circuit without providing power to the second heater wire circuit, and in a second mode, electrical power pass through the first electrical connection to provide power to both the first heater wire circuit and the second heater wire circuit.