Here Global B.v.

A9.com, Inc.

Nokia Technologies Oy

Xi Zhang

Xin Chen

Himanshu Arora

Arnab Sanat Kumar Dhua

Yu Lou

Colin Jon Taylor

David Creighton Mott

Chun-Kai Wang

Sudeshna Pantham

Radek Grzeszczuk

Karl Hillesland

Arnab S Dhua

Andi Zang

Tomas Francisco Yago Vicente

Ting-Hsiang Hwang

Kenan Deng

Sundar Vedula

Images of various views of objects can be captured. An object mesh structure can be created based at least in part on the object images. The object mesh structure represents the three-dimensional shape of the object. Alpha masks indicating which pixels are associated with the object can be used to refine the object mesh structure. A request can be made to view the object from an arbitrary viewpoint which differs from the viewpoints associated with the captured images. A subset of the captured images can be used to create a synthetic image. Different weights can be assigned to the captured image to render a synthetic image that represents the view from the arbitrary viewpoint selected. The input images for the synthetic image can be prefetched, or loaded into memory before the arbitrary view is requested. The images can also be cached for future use or to avoid reloading them for another synthetic image.

Rendering three-dimensional models on mobile devices

A machine learning system receives a reference image and generates a series of projected view images of a physical object represented in the images. Parallel neural networks may receive the reference image and series of projected view images for analysis to determine one or more features of the physical object. By pooling the results from the parallel network, a single output may be provided to a set of decodes that are trained to identify a material property of the one or more items. As a result, a three-dimensional model may be generated that includes a graphical representation of the object as a function of its material properties to enable improved rendering.

Method and system for generating models from multiple views

Systems and methods for displaying 3D containers in a computer generated environment are described. A computing device may provide a user with a catalog of objects which may be purchased. In order to view what an object may look like prior to purchasing the object, a computing device may show a 3D container that has the same dimensions as the object. As discussed herein, the 3D container may be located and oriented based on a two-dimensional marker. Moreover, some 3D containers may contain a representation of an object, which may be a 2D image of the object.

Augmented reality product preview

A method and apparatus for rendering geographic areas involves presenting at least part of the geographic area in a distinctive fashion. An area of interest is identified. A location of the area of interest as a geographic sub-area within a geographic area is determined. A representation of the area of interest within an electronic model of the geographic area is located. A view of the electronic model of the geographic area comprising the representation of the area of interest is selected, and the view of the electronic model with the area of interest having a different display characteristic than other geographic sub-areas shown in the view is presented. The different display characteristic distinguishes the area of interest from the other geographic sub-areas.

Distinguishable geographic area presentation

Structure boundaries may be determined by receiving a plurality of three dimensional (3D) data points representing a geographic area. The 3D data points may be projected into a two dimensional (2D) grid comprised of area elements. A structure boundary may be determined based on an analysis of the area elements.

Structure determination in a geographic area

Systems, apparatuses, and methods are provided for three-dimensional modeling of building roofs using three-dimensional point cloud data. Point cloud data of a roof of a building is received, and roof data points are selected or extracted from the point cloud data. Semantic type classifications are calculated for each selected roof data point. Roof styles are determined from the semantic type classifications, and a synthetic model of the roof and building is rendered based on the determined roof style.

Learning synthetic models for roof style classification using point clouds

Systems, methods, and apparatuses are disclosed for simplifying a point cloud. A point cloud is received where the point cloud has a plurality of points, a global spatial structure, and a local point density. The processor calculates a set of pairwise distances for the plurality of points to at least one other point in the plurality of points. A first distance matrix is generated using the set of pairwise distances. The processor calculates a second pairwise distance set where the plurality of points have a weight and generates a second distance matrix based off the second pairwise distance set. A portion of the points in the second pairwise distance set are removed based on the weight. The processor performs a comparison of the two matrices using the comparison and the global spatial structure and the local point density, and generates a second point cloud based on the second distance matrix.

Structure preserved point cloud simplification

Systems, apparatuses, and methods are provided for refining building alignment in an aerial image. At least one candidate shifting vector and matching score value are determined for a local building. At least one dominant shifting vector is determined for at least one random group of neighboring buildings of the local building. At least one optimized matching score is calculated using the at least one candidate shifting vector for the local building and the at least one dominant shifting vector for the at least one random group of the neighboring buildings. A final shifting vector for the local building is found using the at least one optimized matching score.

Systems and methods for refining building alignment in an aerial image

A method, apparatus and computer program product are provided for roof type classification and reconstruction based on two dimensional aerial images. In the context of a method, the method includes receiving a roof image, determining a segmentation of the roof image based on cutting lines associated with roof features and classifying roof segments based on roof features within the segment. The classifying roof segments is based on the roof features correlation to a roof type pattern.

Method and apparatus for roof type classification and reconstruction based on two dimensional aerial images

A computing device performs matching between a target image and one or more template images. The computing device receives image data and performs an edge detection algorithm on the image data. The edge detection algorithm includes a distance metric based on angles between gradient vectors in the image data and gradient vectors in one or more templates. The computing device matches a building model to the image data based on results of the edge detection algorithm, wherein the building model is associated with the one or more templates.

Growing community of inventors

Chicago, IL, United States of America

Xi Zhang