An image processing method generates a training dataset for training a machine learning model to predict illumination conditions for different positions relative to a scene, the training dataset including training images and reference data. The method includes: obtaining a training image of a training scene acquired by a first camera having an associated first coordinate system; determining local illumination maps associated to a respective position in the training scene in a respective second coordinate system and representing illumination received from different directions around the position; transforming the position of each local illumination map from the second to the first coordinate system; responsive to determining that the transformed position of a local illumination map is visible: transforming the local illumination map from the second to the first coordinate system and including the transformed local illumination map and its transformed position in the reference data associated to the training image.

There is provided a method of re-classifying a clinically significant feature of a medical image as an artifact, comprising: feeding a target medical image captured by a specific medical imaging sensor at a specific setup into a machine learning model, obtaining a target feature map as an outcome of the machine learning model, wherein the target feature map includes target features classified as clinically significant, analyzing the target feature map with respect to sample feature map(s) obtained as an outcome of the machine learning model fed a sample medical image captured by at least one of: the same specific medical imaging sensor and the same specific setup, wherein the sample feature map(s) includes sample features classified as clinically significant, identifying target feature(s) depicted in the target feature map having attributes matching sample feature(s) depicted in the sample feature map(s), and re-classifying the identified target feature(s) as an artifact.

There is provided a method for predicting characteristic information of a target to be recognized. The method comprises: acquiring a plurality of first face images for learning and characteristic information on each first face image; generating a plurality of second face images for learning obtained by synthesizing a mask image with the plurality of first face images for learning by a predetermined algorithm; and training a first neural network by using the plurality of second face images for learning as input data for learning and characteristic information as label data for each second face image corresponding to one of the first face images.

Certain aspects are directed to an apparatus for lane estimation. The apparatus generally includes: at least one memory; and at least one processor coupled to the at least one memory and configured to receive a first input associated with a three-dimensional (3D) space, extract, from the first input, a first set of points associated with a ground plane of the 3D space, map each of the first set of points to a region of a plurality of regions of a two-dimensional (2D) frame, determine one or more attributes associated with each region of the plurality of regions based on one or more of the first set of points mapped to the region, and identify one or more road lanes based on the one or more attributes.

Certain aspects of the present disclosure provide techniques and apparatus for efficient processing of visual content using machine learning models. An example method generally includes generating, from an input, an embedding tensor for the input. The embedding tensor for the input is projected into a reduced-dimensional space projection of the embedding tensor based on a projection matrix. An attention value for the input is derived based on the reduced-dimensional space projection of the embedding tensor and a non-linear attention function. A match, in the reduced-dimensional space, is identified between a portion of the input and a corresponding portion of a target against which the input is evaluated based on the attention value for the input. One or more actions are taken based on identifying the match.

Certain aspects of the present disclosure provide techniques and apparatus for efficient processing of visual content using machine learning models. An example method generally includes generating, from an input, an embedding tensor for the input. The embedding tensor for the input is projected into a reduced-dimensional space projection of the embedding tensor based on a projection matrix. An attention value for the input is derived based on the reduced-dimensional space projection of the embedding tensor and a non-linear attention function. A match, in the reduced-dimensional space, is identified between a portion of the input and a corresponding portion of a target against which the input is evaluated based on the attention value for the input. One or more actions are taken based on identifying the match.

Embodiments of the present disclosure provide a method, device, and computer program product for error evaluation. A method for error evaluation comprises in accordance with a determination that an error occurs in a data protection system, obtaining context information related to an operation of the data protection system; determining, based on the context information and using a trained deep learning model, a type of the error in the data protection system from a plurality of predetermined types, the deep learning model being trained based on training context information and a label on a ground-truth type of an error associated with the training context information; and providing the determined type of the error in the data protection system. In this way, it is possible to achieve automatic classification of errors in the data protection system, thereby improving the efficiency in error classification and saving the operation costs. Therefore, more rapid and more accurate measures can be taken to handle the errors.

A signal translating method may include, according to one aspect of the present application, receiving a source signal of a first domain; identifying erroneous features and effective features from the source signal; translating the source signal of the first domain into a first virtual signal of a second domain, the first virtual signal is that in which erroneous features included in the source signal has been removed; and outputting the first virtual signal. Therefore, the virtual signal of the second domain in which the erroneous features removed may be output.

Various embodiments of a cyber-physical system for providing cloud prediction for photovoltaic array control are disclosed herein.

A dividing pattern determination device capable of reducing the amount of computation performed when determining a dividing pattern of an image. An image for which a dividing pattern is expressed by a hierarchical structure for each predetermined area is input to a feature extraction section, and the feature extraction section generates, based on the input image, for the predetermined area, a hierarchy map in which a value indicative of a block size is associated with each of a plurality of blocks in the predetermined area. A determination section determines a dividing pattern of the image based on the generated hierarchy map.