Discussed herein are systems, devices, and methods for automatic target recognition based on a non-visible input image. A method can include providing, as input to a first machine learning (ML) model for object classification, pixel data of a non-visible image, the first ML model including an encoder from a second ML model, the second ML model trained to generate a visible image representation of an input non-visible image, and receiving, from the first ML model, data indicating one or more objects present in the non-visible image.

A system and method for correlating video frames in a computing environment. The method includes receiving first video data and second video data from one or more data sources. The method further includes encoding the received first video data and the second video data using machine learning network. Further, the method includes generating first embedding video data and second embedding video data corresponding to the received first video data and the received second video data. Additionally, the method includes determining a contrastive IDM temporal regularization value for the first video data and the second video data. The method further includes determining temporal alignment loss between the first video data and the second video data. Also, the method includes determining correlated video frames between the first video data and the second video databased on the determined temporal alignment loss and the determined contrastive IDM temporal regularization value.

The present disclosure provides a method, a device, and a computer program product using a self-supervised anatomical embedding (SAM) method. The method includes randomly selecting a plurality of images; for each image of the plurality of images, performing random data augmentation to obtain a patch pair, generating global and local embedding tensors for each patch of the patch pair, and selecting positive pixel pairs from the patch pair and obtaining positive embedding pairs; for each positive pixel pair, computing global and local similarity maps, finding global hard negative embeddings, selecting global random negative embeddings, pooling the global hard negative embeddings and the global random negative embeddings to obtain final global negative embeddings, and finding local hard negative embeddings using the global and local similarity maps, and randomly sampling final local negative embeddings from the local hard negative embeddings; and minimizing a final info noise contrastive estimation (InfoNCE) loss.

A landmark detection system can more accurately detect landmarks in images using a detection scheme that penalizes for dispersion parameters, such as variance or scale. The landmark detection system can be trained using both labeled and unlabeled training data in a semi-supervised approach. The landmark detection system can further implement tracking of an object across multiple images using landmark data.

Methods and systems are disclosed for improvements in aquaculture that allow for increasing the number and harvesting efficiency of fish in an aquaculture setting by identifying and predicting internal conditions of the juvenile fish based on external characteristics that are imaged through non-invasive means.

Disclosed herein are arrangements that facilitate the transfer of knowledge from models for a source data-processing domain to models for a target data-processing domain. A convolutional neural network space for a source domain is factored into a first classification space and a first reconstruction space. The first classification space stores class information and the first reconstruction space stores domain-specific information. A convolutional neural network space for a target domain is factored into a second classification space and a second reconstruction space. The second classification space stores class information and the second reconstruction space stores domain-specific information. Distribution of the first classification space and the second classification space is aligned.

Techniques for feedback-based training are described. An exemplary method includes receiving a request to perform feedback-based retraining, the request including one or more of an identifier of one or more models to retrain, an identifier of a dataset to use for retraining, an identifier of a dataset to use for testing, an indication of a threshold for an anomaly, an indication of how to display items to verify, and an indication of where to store historical information; applying the selected scoring machine learning model on an unlabeled dataset to generate, per dataset item of the unlabeled dataset, at least one of a score and a confidence for the score; providing a result of the application of the selected scoring machine learning model on an unlabeled dataset to request feedback in the form of a graphical user interface; receiving the requested feedback via the graphical user interface; adding data from the unlabeled dataset into the training dataset when the received requested feedback indicates a verified result; and retraining the selected scoring machine learning model using the training data with the added data from the unlabeled dataset.

A system includes a computing platform having a hardware processor and a memory storing a software code and a neural network (NN) having multiple layers including a last activation layer and a loss layer. The hardware processor executes the software code to identify different combinations of layers for testing the NN, each combination including candidate function(s) for the last activation layer and candidate function(s) for the loss layer. For each different combination, the software code configures the NN based on the combination, inputs, into the configured NN, a training dataset including multiple data objects, receives, from the configured NN, a classification of the data objects, and generates a performance assessment for the combination based on the classification. The software code determines a preferred combination of layers for the NN including selected candidate functions for the last activation layer and the loss layer, based on a comparison of the performance assessments.

A contrastive learning method for color constancy employs a fully-supervised construction of contrastive pairs, driven by a novel data augmentation. The contrastive learning method includes receiving two training images, constructing positive and negative contrastive pairs by the novel data augmentation, extracting representations by a feature extraction function, and training a color constancy model by contrastive learning representations in the positive contrastive pair are closer than representations in the negative contrastive pair. The positive contrastive pair contains images having an identical illuminant while negative contrastive pair contains images having different illuminants. The contrastive learning method improves the performance without additional computational costs. The desired contrastive pairs allow the color constancy model to learn better illuminant feature that are particular robust to worse-cases in data sparse regions.

An automatic labeling method for assigning labels to unlabeled point clouds among a set of labeled and unlabeled point clouds includes preparing an initial machine learning classification model, selecting a labeled point cloud for each of the unlabeled point clouds based on similarities between a feature vector of each of the unlabeled point clouds output through the model and feature vectors of the labeled point clouds output through the model and assigning a cluster label to each of the unlabeled point clouds based on a label of the selected labeled point cloud, assigning pseudo labels to the unlabeled point clouds to which the cluster labels are assigned based on a confidence score obtained through the model, and updating the model by training the model with the labeled point clouds and the unlabeled point clouds to which the pseudo labels are assigned.