Systems and methods are provided for provided for automatic evaluation of sperm morphology. An image of a semen sample is obtained, and at least a portion of the image is provided to a convolutional neural network classifier. The convolutional neural network classifier evaluates the portion of the image to assign to the portion of the image a set of likelihoods that the portion of the image belongs to a plurality of output classes representing the morphology of sperm within the portion of the image. A metric is assigned to the semen sample based on the likelihoods assigned by the convolutional neural network.

In one embodiment, a method for solution inference using neural networks in LiDAR localization includes constructing a cost volume in a solution space for a predicted pose of an autonomous driving vehicle (ADV), the cost volume including a number of sub volumes, each sub volume representing a matching cost between a keypoint from an online point cloud and a corresponding keypoint on a pre-built point cloud map. The method further includes regularizing the cost volume using convention neural networks (CNNs) to refine the matching costs; and inferring, from the regularized cost volume, an optimal offset of the predicted pose. The optimal offset can be used to determining a location of the

A system includes a first neural network configured to calculate, based on input data, data indicative of a predicted result of a predetermined prediction task for the input data, and a second neural network configured to calculate, based on the input data and labelled data corresponding to the input data, data related to error in the labelled data. At least one of the first neural network or the second neural network is trained by using at least both the data indicative of the predicted result calculated by the first neural network and the data related to the error in the labelled data calculated by the second neural network.

An automatic target recognizer system including: a database that stores target recognition data including multiple reference features associated with each of multiple reference targets; a pre-selector that selects a portion of the target recognition data based on a reference gating feature of the multiple reference features; a preprocessor that processes an image received from an image acquisition system which is associated with an acquired target and determines an acquired gating feature of the acquired target; a feature extractor and processor that discriminates the acquired gating feature with the reference gating feature and, if there is a match, extracts multiple segments of the image and detects the presence, absence, probability or likelihood of one of multiple features of each of the multiple reference targets; a classifier that generates a classification decision report based on a determined classification of the acquired target; and a user interface that displays the classification decision report.

A vison-based self-checkout terminal is provided. Purchased items are placed on a base and multiple cameras take multiple images of each item placed on the base. A location for each item placed on the base is determined along with a depth and the dimensions of each item at its given location on the base. Each item's images are then cropped, and item recognition is performed for each item on that item's cropped images with that item's corresponding depth and dimension attributes. An item identifier for each item is obtained along with a corresponding price and a transaction associated with items are completed.

Techniques for determining a classification probability of an object in an environment are discussed herein. Techniques may include analyzing sensor data associated with an environment from a perspective, such as a top-down perspective, using multi-channel data. From this perspective, techniques may determine channels of multi-channel input data and additional feature data. Channels corresponding to spatial features may be included in the multi-channel input data and data corresponding to non-spatial features may be included in the additional feature data. The multi-channel input data may be input to a first portion of a machine-learned (ML) model, and the additional feature data may be concatenated with intermediate output data from the first portion of the ML model, and input into a second portion of the ML model for subsequent processing and to determine the classification probabilities. Additionally, techniques may be performed on a multi-resolution voxel space representing the environment.

Techniques for determining a classification probability of an object in an environment are discussed herein. Techniques may include analyzing sensor data associated with an environment from a perspective, such as a top-down perspective, using multi-channel data. From this perspective, techniques may determine channels of multi-channel input data and additional feature data. Channels corresponding to spatial features may be included in the multi-channel input data and data corresponding to non-spatial features may be included in the additional feature data. The multi-channel input data may be input to a first portion of a machine-learned (ML) model, and the additional feature data may be concatenated with intermediate output data from the first portion of the ML model, and input into a second portion of the ML model for subsequent processing and to determine the classification probabilities. Additionally, techniques may be performed on a multi-resolution voxel space representing the environment.

Systems, methods, apparatuses, and computer program products for detecting, identifying, and monitoring objects. One method may include detecting, by a camera, at least one object according to at least one coordinate; detecting, by a sensor, at least one object according to at least one coordinate; fusing, by a computer vision application, the at least one coordinate of the at least one object detected by the camera with the at least one coordinate of the same at least one object detected by the sensor; determining, by a computer vision application, whether the number of objects detected by the camera equals the number of objects detected by the sensor; and determining, by the computer vision application, whether a score associated with the one or more additional functions is below a predetermined threshold.

An image processing method is provided, including: obtaining a target image; invoking an image recognition model including: a backbone network, a pooling module and a dilated convolution module that are connected to the backbone network and that are parallel to each other, and a fusion module connected to the pooling module and the dilated convolution module; performing feature extraction on the target image by extracting, using the backbone network, a feature map of the target image, separately processing, using the pooling module and the dilated convolution module, the feature map, to obtain a first result outputted by the pooling module and a second result outputted by the dilated convolution module, and fusing the first result and the second result by using the fusion module into a model recognition result of the target image; and determining a semantic segmentation labeled image of the target image based on the model recognition result.

A method for processing breast tissue image data includes processing the image data to generate a set of image slices collectively depicting the patient's breast; for each image slice, applying one or more filters associated with a plurality of multi-level feature modules, each configured to represent and recognize an assigned characteristic or feature of a high-dimensional object; generating at each multi-level feature module a feature map depicting regions of the image slice having the assigned feature; combining the feature maps generated from the plurality of multi-level feature modules into a combined image object map indicating a probability that the high-dimensional object is present at a particular location of the image slice; and creating a 2D synthesized image identifying one or more high-dimensional objects based at least in part on object maps generated for a plurality of image slices.