Provided are a learning data collection apparatus, a learning data collection method, and a program for collecting learning data to be used for efficient retraining. A learning data collection apparatus (10) includes an inspection image acquisition unit (11) that acquires an inspection image, a region detection result acquisition unit (damage detection result acquisition unit (13)) that acquires a region detection result the region detection result indicating a region detected by a region detector that is trained, a correction history acquisition unit (15) that acquires a correction history of the region detection result, a calculation unit (17) that calculates correction quantification information obtained by quantifying the correction history, a database that stores the inspection image, the region detection result, and the correction history in association with each other, an image extraction condition setting unit (19) that sets a threshold value of the correction quantification information as an extraction condition, the extraction condition being a condition for extracting the inspection image to be used for retraining from the database, and a first learning data extraction unit (21) that extracts, as learning data for retraining the region detector, the inspection image satisfying the extraction condition and the region detection result and the correction history that are associated with the inspection image.

A training data creation apparatus, method, and program that can create training data for training a region extractor with the expected performance in a situation in which a plurality of ground-truth region masks have been assigned to a single image, a machine learning apparatus and method, a learning model, and an image processing apparatus are provided. A training data creation apparatus includes a first processor, in which a training sample acquisition unit of first processor acquires a single image and a plurality of first ground-truth region masks for the single image from a database as a single training sample. A ground-truth region mask combination unit generates a single second ground-truth region mask from the first ground-truth region masks included in the training sample. An output unit outputs, as training data, the pair of the single image included in the training sample and the combined second ground-truth region mask.

A neural network is trained to segment interferogram images. A first plurality of interferograms are obtained, where each interferograms corresponds to data acquired by an OCT system using a first scan pattern, annotating each of the plurality of interferograms to indicate a tissue structure of a retina, training a neural network using the plurality of interferograms and the annotations, inputting a second plurality of interferograms corresponding to data acquired by an OCT system using a second scan pattern and obtaining an output of the trained neural network indicating the tissue structure of the retina that was scanned using the second scan pattern. The system and methods may instead receive a plurality of A-scans and output a segmented image corresponding to a plurality of locations along an OCT scan pattern.

A method for machine learning-based classification may include training a machine learning model with a full training data set, the full training data set comprising a plurality of data points, to generate a first model state of the machine learning model, generating respective embeddings for the data points in the full training data set with the first model state of the machine learning model, applying a clustering algorithm to the respective embeddings to generate one or more clusters of the embeddings, identifying outlier embeddings from the one or more clusters of the embeddings, generating a reduced training data set comprising the full training data set less the data points associated with the outlier embeddings, training the machine learning model with the reduced training data set to a second model state, and applying the second model state to one or more data sets to classify the one or more data sets.

In various examples, a background of an object may be modified to generate a training image. A segmentation mask may be generated and used to generate an object image that includes image data representing the object. The object image may be integrated into a different background and used for data augmentation in training a neural network. Data augmentation may also be performed using hue adjustment (e.g., of the object image) and/or rendering three-dimensional capture data that corresponds to the object from selected views. Inference scores may be analyzed to select a background for an image to be included in a training dataset. Backgrounds may be selected and training images may be added to a training dataset iteratively during training (e.g., between epochs). Additionally, early or late fusion nay be employed that uses object mask data to improve inferencing performed by a neural network trained using object mask data.

A computer-program product storing instructions which, when executed by a computer, cause the computer to receive an input data from a sensor, wherein the input data includes data indicative of an image, wherein the sensor includes a video, radar, LiDAR, sound, sonar, ultrasonic, motion, or thermal imaging sensor, generate an adversarial version of the input data, utilizing a generator, in response to the input data, create a training data set utilizing the input data and the adversarial version of the input data, determine an update direction of a meta model utilizing stochastic gradient respect with respect to an adversarial loss, and determine a cross-entropy based classification loss in response to the input data and classification utilizing a classifier, and update the meta model and the classifier in response to the cross-entropy classification loss utilizing the training data set.

Disclosed herein is digital object generator that makes uses a one-way function to generate unique digital objects based on the user specific input. Features of the input are first extracted via a few-shot convolutional neural network model, then evaluated weight and integrated fit. The resulting digital object includes a user decipherable output such as a visual representation, an audio representation, or a multimedia representation that includes recognizable elements from the user specific input.

Methods, systems, an apparatus, including computer programs encoded on a storage device, for training an image classifier. A method includes receiving an image that includes a depiction of an object; generating a set of poorly localized bounding boxes; and generating a set of accurately localized bounding boxes. The method includes training, at a first learning rate and using the poorly localized bounding boxes, an object classifier to classify the object; and training, at a second learning rate that is lower than the first learning rate, and using the accurately localized bounding boxes, the object classifier to classify the object. The method includes receiving a second image that includes a depiction of an object; and providing, to the trained object classifier, the second image. The method includes receiving an indication that the object classifier classified the object in the second image; and performing one or more actions.

Disclosed is a method for recognizing a marine object based on hyperspectral data including collecting target hyperspectral data; preprocessing the target hyperspectral data; and detecting and identifying an object included in the target hyperspectral data based on a marine object detection and identification model, trained through learning of the detection and identification of the marine object. According to the present invention, the preprocessing and processing of the hyperspectral data collected in real time according to a communication state may be performed in the sky or on the ground.

A method is provided to using a machine learning model to predict lane geometry where incorrect or missing lane line geometry is detected. Methods may include: receiving a representation of lane line geometry for one or more roads of a road network; identifying an area within the representation including broken lane line geometry; generating a masked area of the area within the representation including the broken lane line geometry; processing the representation with the masked area through an inpainting model, where the inpainting model includes a generator network, where the representation is processed through the generator network which includes dilated convolution layers for inpainting of the masked area with corrected lane line geometry in a corrected representation; and updating a map database to include the corrected lane line geometry in place of the area including the broken lane line geometry based on the corrected representation.