The described technology is a technique related to an indoor localization method using deep learning. The indoor localization method comprises: opening a 3D tour comprising a plurality of panoramic images; receiving a first perspective image captured by a camera provided in the user device; calculating global features for the first perspective image and each of the plurality of panoramic images included in the 3D tour; selecting a most similar panoramic image to the first perspective image by using the calculated global features; computing an indoor location corresponding to a location of the camera on the 3D tour by using feature points included in the selected panoramic image and the first perspective image; and providing the computed indoor location to the user device.

A smartphone may be freely moved in three dimensions as it captures a stream of images of an object. Multiple image frames may be captured in different orientations and distances from the object and combined into a composite image representing an image of the object. The image frames may be formed into the composite image based on representing features of each image frame as a set of points in a three dimensional point cloud. Inconsistencies between the image frames may be adjusted when projecting respective points in the point cloud into the composite image. Quality of the image frames may be improved by processing the image frames to correct errors. Further, operating conditions may be selected, automatically or based on instructions provided to a user, to reduce motion blur. Techniques, including relocalization such that, allow for user-selected regions of the composite image to be changed.

The present disclosure provides a laser sensor-based map generation method. In an embodiment, the method includes: obtaining image data, the image data being acquired by a visual sensor; determining first point cloud data belonging to glass-like region in laser data based on the image data; adjusting a weight of the laser data according to the first point cloud data; fusing the first point cloud data and second point cloud data belonging to non-glass-like region in the laser data based on the adjusted weight of the laser data, to generate a map.

An information processing apparatus includes: a memory; and a processor coupled to the memory and configured to: calculate a positional difference and a degree of similarity in feature value between each feature point in a verification image captured by an imaging device and each of a plurality of feature points in a registered image registered in advance; and match the verification image against the registered image by using the degree of similarity to a feature point having the degree of similarity that is second highest or lower for each feature point in the registered image in a case where a positional difference from a feature point having the degree of similarity that is highest in the verification image is greater than a threshold value.

A method for generating and evaluating N-to-1 mappings between spatial point sets in nD, n=2 or 3, implemented on a computing device comprising a programmable general purpose processor and a programmable data-parallel coprocessor and a memory coupled with them. Embodiments of the method comprises using the computing device to carry out steps comprising receiving a first and a second spatial point sets in 2D or 3D, the first spatial point set comprising a first non-empty non-isolated portion of non-isolated points and a second constrained portion of constrained points, receiving an extended array of fixed correspondents comprising a first not-yet-fixed portion for the non-isolated portion of the first spatial point set and a second fixed portion for the constrained portion of the first spatial point set, a CCISS or padded CCISS between the first non-empty non-isolated portion and the second spatial point set, dividing the first non-empty non-isolated portion into a number of sub-portions, and dividing the first not-yet-fixed portion of the extended array of fixed correspondents and the CCISS or the padded CCISS accordingly, iteratively generating optimal N-to-1 mappings between the members of the sub-portions of the first non-empty on-isolated portion and updating the respective sub-portions of the extended array of fixed correspondents one sub-portion at each iteration.

The present disclosure provides a method for evaluating an image acquisition accuracy of a Demura device, including: controlling a display panel to display a detection picture, wherein the detection picture includes a plurality of test point patterns with an interval therebetween; acquiring an image of the detection picture by the Demura device to obtain a preprocessed image corresponding to the detection picture, wherein the preprocessed image and a corresponding detection picture have a same size and a same shape; and determining the image acquisition accuracy of the Demura device according to a difference between a position of each of the plurality of test point patterns in the detection picture and a corresponding position of the test point pattern in the preprocessed image. The present disclosure also provides an apparatus for evaluating an image acquisition accuracy of a Demura device, an electronic device and a non-transitory computer-readable storage medium.

A method for ascertaining features in an environment of at least one mobile unit for implementation of a localization and/or mapping by a control unit. In the course of the method, sensor measurement data of the environment are received, the sensor measurement data received are transformed by an alignment algorithm into a cost function and a cost map is generated with the aid of the cost function, a convergence map is generated based on the alignment algorithm. At least one feature is extracted from the cost map and/or the convergence map and stored, the at least one feature being provided in order to optimize a localization and/or mapping. A control unit, a computer program, and a machine-readable storage medium are also described.

Examples described herein include methods for producing an augmented reality image. An example method includes a portable computer terminal obtaining a first estimated position and a first estimated orientation of a first scene; producing a first estimated image of the first scene; obtaining a first camera picture of the first scene; executing a first pattern matching between the first estimated image and the first camera picture; adjusting the approximate position and/or orientation of the portable computer terminal; executing a second pattern matching; and producing an augmented reality image considering the current position and orientation of the portable computer terminal.

A system for object detection and posture estimation based on an object-customized image feature detection algorithm according to an embodiment of the present disclosure comprises at least one or more processors; and at least one or more memories, wherein at least one application, as an application that is stored in the memory and performs object detection and posture estimation based on an object-customized image feature detection algorithm by being executed by the at least one or more processors, learns a goal object based on a plurality of image feature detection algorithms, generates an algorithm list that includes evaluation of the plurality of image feature detection algorithms for detecting the learned goal object, detects a target object corresponding to the goal object among at least one or more candidate objects within an input image based on the generated algorithm list, and performs a posture estimation process for the detected target object based on the generated algorithm list.

A computer-implemented method to generate instructions for a computing device. A first graph having nodes and edges is provided, which defines first instructions for the computing device. At least one first part is sought in the first graph. A second part is determined as a function of the at least one first part. A directed, acyclic, linked second graph having nodes and edges is determined as a function of the first graph. In the second graph, the first part is replaced by the second part. The second graph defines second instructions for the computing device for executing the computational algorithm. A pattern for at least a part of a graph is provided, whose nodes and edges are defined by instructions that are executable by the computing device. The first graph or the second graph is selected as a function of the pattern, to generate instructions for the computing device.