The present disclosure provides a tracking control method. The method includes obtaining an input image sequence; based on a detection algorithm, detecting a frame of input image in the input image sequence to obtain a tracking frame including a target object; and based on a tracking algorithm, tracking the target object in a plurality of frames of input images behind the frame of input image based on the tracking frame of the target object.

Aspects of the disclosure provide for mechanisms for identification of fields in documents using neural networks. A method of the disclosure includes obtaining a layout of a document, the document having a plurality of fields, identifying the document, based on the layout, as belonging to a first type of documents of a plurality of identified types of documents, identifying a plurality of symbol sequences of the document, and processing, by a processing device, the plurality of symbol sequences of the document using a first neural network associated with the first type of documents to determine an association of a first field of the plurality of fields with a first symbol sequence of the plurality of symbol sequences of the document.

Active sensor fusion systems and methods may include a plurality of sensors, a plurality of detection algorithms, and an active sensor fusion algorithm. Based on detection hypotheses received from the plurality of detection algorithms, the active sensor fusion algorithm may instruct or direct modifications to one or more of the plurality of sensors or the plurality of detection algorithms. In this manner, operations of the plurality of sensors or processing of the plurality of detection algorithms may be refined or adjusted to provide improved object detection with greater accuracy, speed, and reliability.

Machine learning based models recognize objects in images. Specific features of the object are extracted from the image using machine learning based models. The specific features extracted from the image assist deep learning based models in identifying subtypes of a type of object. The system recognizes the objects and collections of objects and determines whether the arrangement of objects violates any predetermined policies. For example, a policy may specify relative positions of different types of objects, height above ground at which certain types of objects are placed, or an expected number of certain types of objects in a collection.

A vehicle accident detection method and system is provided. The method includes receiving location coordinates associated with a location of an occurring vehicular accident. Data associated with possible causes of the vehicular accident is received from sensors. Traffic related rules associated with a geographical location are retrieved and analyzed with respect to the data. Parameters associated with at least one vehicle involved in the vehicular accident and a possible cause are determined via execution of programming logic and transmitted to additional systems. The possible cause for the vehicular accident is determined from all possible causes based on matching current and historical accident circumstances. Additionally, weighting factors may be available and adjusted over time for accurate accident detection. A possible cause comprising a greatest weighting factor may be used to identify a most likely cause.

A system and method for model auto-selection for a prediction using an ensemble of machine learning models. The method includes: receiving historical data, the historical data including previous outcomes of a plurality of events associated with a plurality of data categories; training candidate machine learning models with the historical data, each candidate machine learning model trained using a respective one of the data categories; and determining an ensemble of machine learning models by determining a median prediction for combinations of candidate machine learning models and determining the combination that has the median prediction that is closest to at least one of the previous outcomes.

A system for preventing a false alarm that occurs at a building, the system includes a processing circuit configured to receive, via a communications interface, building data including events for the building devices. The processing circuit is configured to determine, based on the events, whether a false alarm rule has triggered, where the false alarm rule indicates relationships between one or more of the events that is indicative of a situation at the building site that causes the false alarm, generate a parameter update for at least one of the plurality of building devices in response to determining that the false alarm rule has triggered, and implement the parameter update by providing, via the communications interface, the parameter update to the at least one of the building devices.

A computer-implemented method of providing a recommendation as to a fault determination for a motor vehicle collision is disclosed. The method may include receiving unstructured text describing the circumstances of the collision. The unstructured text is evaluated an associated intent related to the circumstances of the motor vehicle collision is identified. The intent is mapped to an internal node of a decision tree corresponding to a set of fault-determination rules. The computer then successively prompts and receive input responsive to the prompting that corresponds to details of the circumstances of the collision. The computer may identify, based on the received input, a path through the decision tree ending at a leaf node that corresponds to a fault-determination rule governing motor vehicle collisions that matches the circumstances of the motor vehicle collision. The recommendation is then provided based on that rule. Related systems and computer-readable media are also disclosed.

A computing method receives a labeled sample from an annotator. The method may determine a plurality of reference model risk scores for the first labeled sample, where each reference model risk score corresponds to an amount of risk associated with adding the first labeled sample to a respective reference model of a plurality of reference models. The method may determine an overall risk score for the first labeled sample based on the plurality of reference model risk scores. The method may further determine a probe for confirmation of the first labeled sample and a trust score for the annotator by sending the probe to one or more annotators. In response to determining a trust score for the annotator the method may add the labeled sample to a ground truth or reject the labeled sample.

To prevent intent classifiers from potentially choosing intents that are ineligible for the current input due to policies, dynamic intent classification systems and methods are provided that dynamically control the possible set of intents using environment variables (also referred to as external variables). Associations between environment variables and ineligible intents, referred to as culling rules, are used.