An approach is provided for filtering imagery to train a feature detection model. The approach involves, for example, receiving a plurality of images. The plurality of images is classified as depicting a feature of interest. The approach also involves providing data for presenting a bulk arrangement of at least one subset of the plurality of images, wherein the bulk arrangement is based on a characteristic of the plurality of images (e.g., detection confidence, position, size, visual characteristic, etc. of the detected feature). The approach further involves initiating a filtering of the plurality of images based on the bulk arrangement and providing the filtered plurality of images as training data to train the feature detection model.

Aspects of the disclosure relate to monitoring devices at enterprise locations using machine-learning models to protect enterprise-managed information and resources. In some embodiments, a computing platform may receive, from one or more data source computer systems, passive monitoring data. Based on applying a machine-learning classification model to the passive monitoring data received from the one or more data source computer systems, the computing platform may determine to trigger a data capture process at an enterprise center. In response to determining to trigger the data capture process, the computing platform may initiate an active monitoring process to capture event data at the enterprise center. Thereafter, the computing platform may generate one or more alert messages based on the event data captured at the enterprise center. Then, the computing platform may send the one or more alert messages to one or more enterprise computer systems.

A method, computer program, and computer system is provided for estimating three-dimensional hand poses in images. Data corresponding to two hand images is receive, and an optical flow value corresponding to a change in a hand gesture in the received hand image data is calculate. A heat map is generated based on the calculated optical flow, and a hand mesh map is estimated based on the generated heat map. A hand pose present within the hand images is determined based on the estimated hand mesh map.

Method and apparatus that includes receiving a query describing an aspect in a video, the video including a plurality of frames, identifying multiple proposals that potentially correspond to the query where each of the proposals includes a subset of the plurality of frames, ranking the proposals using a graph convolution network that identifies relationships between the proposals, and selecting, based on the ranking, one of the proposals as a video segment that correlates to the query.

A method for generating an event, which is performed by a doorbell camera system disclosed in the present disclosure may include acquiring video of a visitor, outputting the acquired video of the visitor to a user, learning the user's feedback on the visitor, storing information of the visitor and a learning result of the user's feedback on the visitor, and generating an event based on the information of the visitor and the learning result of the user's feedback.

Pathologists are adopting digital pathology for diagnosis, using whole slide images (WSIs). Explainable AI (xAI) is a new approach to AI that can reveal underlying reasons for its results. As such, xAI can promote safety, reliability, and accountability of machine learning for critical tasks such as pathology diagnosis. HistoMapr provides intelligent xAI guides for pathologists to improve the efficiency and accuracy of pathological diagnoses. HistoMapr can previews entire pathology cases' WSIs, identifies key diagnostic regions of interest (ROIs), determines one or more conditions associated with each ROI, provisionally labels each ROI with the identified conditions, and can triages them. The ROIs are presented to the pathologist in an interactive, explainable fashion for rapid interpretation. The pathologist can be in control and can access xAI analysis via a “why?” interface. HistoMapr can track the pathologist's decisions and assemble a pathology report using suggested, standardized terminology.

The invention is concerned with a method for generating labeling data (27) that describe an image content of images (16) depicting at least one scene (12), wherein in a processing device (18) image data (15) are received from a imaging (14) and a segmentation unit (19) detects at least one object (17) in the image data (15) and a graphical processing unit (20) generates a respective graphical object marker (29) that marks the at least one detected object (17) and a display control unit (21) displays an overlay (35) of the at least one scene (12) and the at least one object marker (29) and an input reception unit (22) receives a respective user input (36) for each object marker (29), wherein the respective user input (36) provides the image content of the image region (26) marked by the object marker (29).

Methods and systems for a monitoring system for data collection in an industrial environment including a data collector communicatively coupled to a plurality of input channels connected to data collection points related to machine components, wherein at least one of the plurality of input channels is connected to a data collection point on a rotating machine component; a data acquisition circuit structured to interpret a plurality of detection values from the collected data, each of the plurality of detection values corresponding to at least one of the plurality of input channels; and an expert system analysis circuit structured to analyze the collected data, wherein the expert system analysis circuit determines a failure state for the rotating machine component based on analysis of the plurality of detection values, wherein upon determining the failure state the expert system analysis circuit provides the failure state to a data storage.

Vacuum seal packages can be classified based on image data. Training image data is received that includes image data about first vacuum seal packages. Labels associated with the first vacuum seal packages are received, where each of the labels includes a state of one of the first vacuum seal packages. A trained classification model is developed based on the training image data and the received labels. Image data representative of a second vacuum seal package is received. The image data is inputted into the trained classification model, where the trained classification model is configured to classify a state of the second vacuum seal package based on the image data. The state of the second vacuum seal package is received from the trained classification model.

Systems, methods, and apparatuses for data collection in a mining environment are disclosed. One exemplary embodiment is a system comprising a data storage structured to store at least one collector route and at least one sensor specification; a data collector communicatively coupled to input channels, and providing detection values from the input channels in response to a selected one of each of the at least one collector route and the at least one sensor specification; a data acquisition circuit structured to interpret the detection values from the data collector; a data analysis circuit structured to: analyze detection values; and determine a data collection quality parameter by evaluating at least one of: the at least one selected sensor collector route and the at least one selected sensor specification; and an analysis response circuit structured to adjust at least one of: the selected at least one sensor collector route and the at least one selected sensor specification, in response to the data collection quality parameter.