Systems and methods herein describe generating a mixture of experts (MoE) models for image classification. The systems and methods include training a plurality of neural network models as experts, wherein the experts are trained to predict an image class, to predict amenities present in the image, to predict location categories in the image, or a combination thereof. The system and methods additionally include training experts based on input differentiation. The system and methods also include training experts having different model architectures or variants of model architectures, and combining the trained experts into an ensemble model. The ensemble model can then be used to classify new images.

A method can include receiving labeled images; acquiring unlabeled images; performing active learning by training an inspection learner using at least a portion of the labeled images to generate a trained inspection learner that outputs information responsive to receipt of one of the unlabeled images by the trained inspection learner; based at least in part on the information, making a decision to call for labeling of the one of the unlabeled images; receiving a label for the one of the unlabeled images; and further training the inspection learner using the label.

A computing device obtains information about a medical slide image, and determines a dataset type of the medical slide image and a panel of the medical slide image. The computing device assigns to an annotator account, an annotation job defined by at least the medical slide image, the determined dataset type, an annotation task, and a patch that is a partial area of the medical slide image. The annotation task includes the determined panel, and the panel is designated as one of a plurality of panels including a cell panel, a tissue panel, and a structure panel. The dataset type indicates a use of the medical slide image and is designated as one of a plurality of uses including a training use of a medical learning model and a validation use of the machine learning model.

An object classification method includes: acquiring image data of an image including feature information indicating a feature of an object; and classifying the object included in the image, based on the feature information. The image data is acquired by causing a first image capture device to capture the image. The first image capture device includes: an image sensor; and a filter array that is arranged on an optical path of light that is incident on the image sensor and that includes translucent filters two-dimensionally arrayed along a plane that crosses the optical path, the translucent filters including two or more filters in which wavelength dependencies of light transmittances are different from each other, and light transmittance of each of the two or more filters having local maximum values in a plurality of wavelength ranges.

A system for broad area geospatial object detection includes a processor configured to retrieve training data including a first plurality of orthorectified geospatial training images each including at least one labeled instance of the object of interest, and a second plurality of orthorectified geospatial images each including at least one labeled instance of the object of interest and/or at least one unlabeled instance of the object of interest, and apply at least one type of image correction to the training data. The processor is also configured to train a plurality of machine learning classifier elements, based on the first plurality of orthorectified geospatial training images and subsequently based on the second plurality of orthorectified geospatial images, each of the plurality of machine learning classifier elements being defined by a machine learning protocol parameterized based on one or more visually unique features of the object of interest.

A method or system uses de-identified images collected from patients in association with de-identified data collected as part of medical care to train machine learning, machine vision, deep learning, or other algorithms that associate an outcome variable with an input image or video image(s).

A method of image processing includes receiving a set of images from a sensor, dynamically determining respective cell resolutions of respective cells in a bird's-eye-view (BEV) grid based on image content in the set of images, wherein at least two of the cells have different cell resolutions, and generating BEV image content based on the respective cell resolutions of the respective cells.

A framework for interpreting machine learning models is proposed that utilizes interpretability methods to determine the contribution of groups of input variables to the output of the model. Input variables are grouped based on dependencies with other input variables. The groups are identified by processing a training data set with a clustering algorithm. Once the groups of input variables are defined, scores related to each group of input variables for a given instance of the input vector processed by the model are calculated according to one or more algorithms. The algorithms can utilize group Partial Dependence Plot (PDP) values, Shapley Additive Explanations (SHAP) values, and Banzhaf values, and their extensions among others, and a score for each group can be calculated for a given instance of an input vector per group. These scores can then be sorted, ranked, and then combined into one hybrid ranking.

A system and method for training a text report identification machine learning model and an image identification machine learning model, including: initially training a text report machine learning model, using a labeled set of text reports including text pre-processing the text report and extracting features from the pre-processed text report, wherein the extracted features are input into the text report machine learning model; initially training an image machine learning model, using a labeled set of images; applying the initially trained text report machine learning model to a first set of unlabeled text reports with associated images to label the associated images; selecting a first portion of labeled associated images; re-training the image machine learning model using the selected first portion of labeled associated images; applying the initially trained image machine learning model to a first set of unlabeled images with associated text reports to label the associated text reports; selecting a first portion of labeled associated text reports; and re-training the text report machine learning model using the selected first portion of labeled associated text reports.

A method, computer system, and a computer program product are provided. A visual inspection machine learning model is trained using a generative adversarial network. Within the generative adversarial network a vector bypass is implemented. By transmitting a vector embedding representation of an unlabeled image through the vector bypass, the vector embedding representation is transmitted around the visual inspection machine learning model and to a generator to assist with image reconstruction.