This disclosure relates generally to a system and method to identify various products on a plurality of images of various shelves of a retail store to facilitate compliance with respect to planograms. Planogram is a visual plan, which designates the placement of products on shelves and merchandising display fixtures of a retail store. Planograms are used to create consistency between store locations, to provide proper shelf space allocation, to improve visual merchandising appeal, and to create product-pairing suggestions. There are a few assumptions considering one instance per product class is available beforehand and the physical dimension of each product template is available in some suitable unit of length. In case of absence of physical dimension of the products, a context information of the retail store will be used. The context information is that the products of similar shapes or classes are arranged together in the shelves for consumers' convenience.

An image matching method includes extracting, from a first image of an object, a landmark patch including a landmark point of the object; extracting, from a second image of the object, a target patch corresponding to the landmark patch; and determining a target point in the second image corresponding to the landmark point based on a matching between the landmark patch and the target patch.

The presented invention includes the generation of cloud points, the identification of objects in the cloud points, and, in this case, finding the positions of objects in cloud points. In addition, the invention includes capturing images, data streaming, and digital image processing in different points of the system, and calculation of the position of objects. The invention includes the usage of cameras of mobile smart devices, smart glasses, 3D cameras, but not necessarily. The data streaming provides video streaming and sensor data streaming from mobile smart devices. The presented invention further includes cloud points of buildings in which the positioning of separated objects could be implemented. It also consists of the database of cloud points of isolated objects which help to calculate the position in the building. Finally, the invention comprises the method of objects feature extraction, comparing in the cloud points and position calculation.

Method for estimating eye-related geometric parameters of a user, comprising the steps of: a. retrieving one input image observation corresponding to an image of the eye; b. using a learning machine for computing a plurality of image segmentation maps, so as to classify each pixel into one eye region; c. generating through a set of geometric parameters an image geometric model of the user's eye; d. comparing the image geometric model with an image segmentation map; e. computing a model correspondence value indicating if said input image observation corresponds to the geometric model; f. repeating steps c. to e. if the value computed under step e. is below an optimal value wherein one parameter is changed for each iteration until said model correspondence value reaches the optimal value, and g. retrieving the eye-related geometric parameters from the latest model of the user's eye.

Provided are VR image processing method and apparatus. The method includes: rendering left-eye and right-eye viewpoint regions based on left-eye and right-eye view angles respectively, to obtain left-eye and right-eye viewpoint images; determining a candidate region based on positions of the left-eye and right-eye view angles, and selecting a point in the candidate region as a peripheral image view angle; rendering left-eye and right-eye viewpoint peripheral regions based on the peripheral image view angle, to obtain a same viewpoint peripheral image; and splicing the viewpoint peripheral image with the left-eye viewpoint image and with the right-eye viewpoint image to obtain a left-eye complete image and a right-eye complete image.

An image scanning device scans one or plural document images of one or plural predetermined head pages, and a preview processing unit performs preview display of the document image(s) of the head page(s). After detecting a user operation to accept the document image on the preview display, the image scanning device scans a document image of a subsequent page to the head page(s), and the image processing unit performs the image process for the document image of the subsequent page. Further, the image processing unit performs a part or whole of the image process for the document image(s) of the head page(s) before the preview display. If a part of the image process is displayed before the preview display, the image processing unit performs a remaining part of the image process after detecting a user operation to accept the document image on the preview display.

Approaches presented herein provide for semantic data matching, as may be useful for selecting data from a large unlabeled dataset to train a neural network. For an object detection use case, such a process can identify images within an unlabeled set even when an object of interest represents a relatively small portion of an image or there are many other objects in the image. A query image can be processed to extract image features or feature maps from only one or more regions of interest in that image, as may correspond to objects of interest. These features are compared with images in an unlabeled dataset, with similarity scores being calculated between the features of the region(s) of interest and individual images in the unlabeled set. One or more highest scored images can be selected as training images showing objects that are semantically similar to the object in the query image.

A relocation method and a mobile machine using the same are provided. The method includes: obtaining a global map and a current scan map of a target scene where a mobile machine is located, and generating a local sub-map based on the global map; obtaining a black boundary in the local sub-map, determining a length and a curve complexity of the black boundary, and determining a weight of the black boundary based on the length and the curve complexity of the black boundary; determining an estimated pose and a target black boundary based on the local sub-map and the current scan image, and obtaining a matching value between the current scan image and the local sub-map based on a weight of the target black boundary; and determining the estimated pose as a relocated pose of the mobile machine in response to the matching value being larger than a preset threshold.

A detection apparatus to extract features from an image; determine the number of candidate regions of the object in the image based on the extracted features, wherein the determined number of the candidate regions is decided by a position and shape of the candidate regions; and to detect the object from the image based on at least the extracted features and the determined number, position and shape of the candidate regions.

A method includes acquiring information on an in-image coordinate point of a region of interest contained in each of a plurality of images respectively photographed by a plurality of image modules; specifying, with reference to information on a position where at least one of the plurality of image modules is installed and information on an in-image coordinate point of a target region of interest contained in an image photographed by the at least one image module, a candidate figure containing a position where the target region of interest is located in a reference space; and specifying the position where the target region of interest is located in the reference space, with reference to a positional relationship between a first candidate figure of the target region of interest corresponding to a first image module and a second candidate figure of the target region of interest corresponding to a second image module.