A database construction method includes determining a target image set that meets a preset image coincidence degree requirement, obtaining a scenario key point information set based on the target image set and natural condition information corresponding to each image, determining, in the scenario key point information set, first scenario key point information corresponding to a scenario key point whose life value on a single mobile terminal is greater than a first preset life value threshold, determining, in the first scenario key point information, second scenario key point information corresponding to a scenario key point whose life value on a plurality of mobile terminals is greater than a second preset life value threshold, and when the second scenario key point information does not match preset scenario key point information in a database, constructing the database based on the second scenario key point information.

A method, system and computer-readable storage medium for performing a counterfactual generation operation. The counterfactual generation operation includes: receiving a subject data point; classifying the data point via a trained classifier, the classifying providing a classified data point; identifying a counterfactual using the classified data point, the counterfactual comprising another datapoint, the another data point being close to the subject data point, the another data point resulting in production of a different outcome when provided to a model when compared to an outcome resulting from the subject data point being provided to the model; and, providing the counterfactual to a destination.

A key point correction apparatus detects a plurality of key points of a target object from each of a plurality of captured images. The key point correction apparatus decides whether or not a plurality of key points of the target object detected from the captured image satisfy an appropriacy condition. When the key points of the target object satisfy the appropriacy condition, the key point correction apparatus determines a correction parameter regarding the target object using the key points. When a correction parameter has been determined for the target object detected from the captured image, the key point correction apparatus corrects the key points of the target object using a correction parameter.

This disclosure provides for methods and a system for multiple instance segmentation and tracking. According to an aspect a method is provided. The method includes sending an image to a backbone network and generating image feature outputs. The method further includes sending the image feature outputs to a spatial attention module for generating a feature map associated with objects in the image. The method further includes sending the feature map to a category feature module for generating an instance category output indicating the objects. The method further includes sending the image feature outputs to a mask generating module for generating masks. The method further includes generating: the instance category output via the category feature module, and the masks via the mask generating module. In some embodiments, the method further includes generating re-identification embedding information associated with the objects based on image feature outputs.

The present invention relates to a detection method for an assembly body multi-view change based on feature matching, comprising the following steps: S1, acquiring a first image and a second image; S2, performing feature point extraction and feature matching on the first image and the second image to obtain a matching pair set, a first unmatched point set of the first image and a second unmatched point set of the second image; S3, acquiring a first to-be-matched area set of the first image according to the first unmatched point set; acquiring a second to-be-matched area set of the second image according to the second unmatched point set; S4, performing feature matching on each first unmatched area and each second unmatched area one by one to obtain a plurality of matching results; and S5, outputting the assembly body change type according to the plurality of matching results.

Systems and methods for generating a three-dimensional reconstruction from a two-dimensional image of an asset are described. In some aspects, at least one computer hardware processor is used to perform: receiving a two-dimensional input image of an asset; encoding, using a first statistical model, the input image into a latent vector; and generating, using a second statistical model, at least one depth image from the latent vector, wherein pixel values in the at least one depth image correspond to coordinates of a point cloud.

A method for training a machine learning model for generating descriptor images for images of one or of multiple objects. The method includes: formation of pairs of images which show the one or the multiple objects from different perspectives; generation, for each image pair, using the machine learning model, of a first descriptor image for the first image, and of a second descriptor image for the second image, which assigns descriptors to points of the one or multiple objects shown in the second image; sampling, for each image pair, of descriptor pairs, which include in each case a first descriptor from the first descriptor image and a second descriptor from the second descriptor image, which are assigned to the same point, and the adaptation of the machine learning method for reducing a loss.

Provided is a method for selecting key points for real-time tracking in a mobile environment, which is used by a key point application executed by at least one processor of a computing device to select key points for real-time tracking in a mobile environment. The method comprises obtaining a target object image capturing a target object; extracting a plurality of temporary key points from the obtained target object image; determining confirmed key points by filtering the plurality of extracted temporary key points according to a predetermined distribution criterion; setting the confirmed key points as final key points for the target object image; storing target object tracking information including the final key points set for the target object image; and providing the target object tracking information to an application executing a predetermined function through the target object tracking in a mobile environment.

A system and method for visually comparing one or more attributes of two fashion products on an e-commerce platform is presented. The system includes an image database, a selection module and a comparison module. The image database includes a plurality of two-dimensional (2D) images captured at the same scale for a plurality of fashion products. The selection module is configured to allow a user to select a target fashion product and a reference fashion product. The comparison module is configured to access a 2D image of the target fashion product and a 2D image of the reference fashion product from the image database; and allow the user to visually compare the target fashion product with the reference fashion product with respect to the one or more attributes by positioning the 2D image of the target fashion product on top or bottom of the 2D image of the reference fashion product.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, to generate a scannable document. In one aspect, a method includes receiving a scan request, wherein the scan request includes a plurality of text images; for each text image of the plurality of text images: rectifying the text image to generate a text image with parallel image lines, generating a plurality of word bounding boxes that enclose one or more connected components in the text image, wherein each word bounding box is associated with a respective word, and generating, for each respective word in the text image, a plurality of points that represent the respective word; combining the plurality of text images to form a single text document; and providing the combined image as a scannable document.