Abstract: Provided are computer-implemented technologies of identifying visual pollutions. The technologies include training expert models based on an enhanced knowledge distillation paradigm in which a student model learns from a number of teacher models via a customized training approached specifically for achieving efficiency and effectiveness, quality-controlling and fine-tuning the trained expert model in a production environment via continuous training under newly incorporated training data and object classifications and factoring feedbacks of the detection result of the model on new training data and/or new object classifications, deploying and applying the expert model in detecting, tracking, logging, counting, and reporting a set of visual pollution elements in an environment where visual pollutions are to be detected.

Abstract: A dental clamp retractor includes tongue and cheek retractor portions separated by a clamp bow having an arcuate shape. A buccal extension arm extends laterally outward from the clamp bow to the cheek retractor portion. A buccal gauze holder is connected to the clamp bow by a flexible arm. The tongue retractor portion connects to the clamp bow by a lingual extension arm. A lingual gauze holder is connected to the clamp bow by a flexible arm. The cheek retractor portion may have a larger surface area than the tongue retractor portion and be formed of an elastomer. The buccal gauze holder and lingual gauze holder may be formed by clips. A dental procedure includes applying the dental clamp retractor to a patient to retract the tongue and cheek and hold gauze on a buccal and/or labial side of a tooth surface of the patient.

Abstract: Provided is an edge-device based system and method for real-time detection and dimension estimation of signages by capturing images thereof by using a mounted monocular camera then processing them using a multi-module system that comprises Object Detection, Segmentation, Tracking, and Dimension Estimation modules to detect, segment, track, and estimate dimension of shop boards. The system/method also applies a deep neural network that is trained with labeled data to optimize the model parameters gathered by capturing video from moving vehicle passing through multiple different shops with unconfined locations.

Abstract: An apparatus for generating an inspection schedule includes one or more processors configured to: receive one or more schedule requests comprising inspector features, establishment features, and penalty features; generate inspector information for an inspector model based on the inspector features; generate establishment information, comprising a probability of imposing a penalty on an establishment of the establishment features, for an establishment model based on the establishment features and the penalty features; and generate, for an inspector of the inspector features using a violation model, the inspection schedule based on the inspector information and the establishment information.

Abstract: An Edge encapsulated method and system for real-time detection of a double parking vehicle blocking parked vehicles by capturing images using a mounted monocular camera then processing using two Semi-Supervised Object Detection (sSOD) stages, that is trained with partly labeled and mostly unlabeled data to optimize the model parameters for the detection problem, gathered by capturing video from moving vehicle passing through multiple different vehicle types and parking systems with unconfined location.

Abstract: The invention pertains to a valve monitor system comprising a valve operation mode selector indicator that provides a local indicator to maintenance personnel of whether a valve operation mode selector switch is placed in a local or remote position. The valve monitor system also provides a circuit to a central control room or facility, to allow an operator there to know whether a pipeline valve controlled by the valve operation mode selector switch is configured for local or remote operation.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for extracting facts from a collection of documents. One of the methods includes obtaining a plurality of seed facts; generating a plurality of patterns from the seed facts, wherein each of the plurality of patterns is a dependency pattern generated from a dependency parse; applying the patterns to documents in a collection of documents to extract a plurality of candidate additional facts from the collection of documents; and selecting one or more additional facts from the plurality of candidate additional facts.

Abstract: A tactile button (10) for activating a tact switch disposed on electronic circuitry enclosed within a housing element (102) having a shoulder slot (106). The shoulder slot (106) is integrally disposed with a shoulder ledge (108) with two pivot slots (110,112) and an underside flange (114). The shoulder member (104) couples to the shoulder slot (106) and comprises an actuating stub (202), a second latch (204), and two pivot protrusions (208,208). The tact switch is activated by the actuating stub (202) when depressing on an outer surface (216) of the shoulder member (104) and movement of the shoulder member (104) is restricted to an angular direction (304) that pivots on the pivot slots (110,112). In an alternate embodiment, a portable electronic device (40) has a left tactile button (402) and a right tactile button (404) and inadvertent activation of either tactile button is prevented because activation of either tactile button (402,404) is not laterally but from a top portion of the outer surface (216).