Microsoft Technology Licensing, LLC

Don Dongwoo Kim

Jonathan Paul Rosenberg

Samuel Thomas Williams

Sandeep Kanumuri

Michael Ebstyne

Emanuel Shalev

Johannes P Kopf

Trebor Lee Connell

Jett Evan Jones

Peter G Sibley

Johanness Kopf

Data representing a scene is received. The scene includes labeled elements such as walls, a floor, a ceiling, and objects placed at various locations in the scene. The original received scene may be modified in different ways to create new scenes that are based on the original scene. These modifications include adding clutter to the scene, moving one or more elements of the scene, swapping one or more elements of the scene with different labeled elements, changing the size, color, or materials associated with one or more of the elements of the scene, and changing the lighting used in the scene. Each new scene may be used to generate labeled training data for a classifier by placing a virtual sensor (e.g., a camera) in the new scene, and generating sensor output data for the virtual sensor based on its placement in the new scene.

Generating training data for machine learning classifier

One or more techniques and/or systems are disclosed for generating improved planar panorama imagery, which may be used to view street-level, human-scale locations along a desired path, for example. One or more desired objects can be detected in source imagery that may be used to produce the resulting planar panorama imagery. Respective object regions comprising a desired object can be identified in one or more frames of the source imagery. One or more portions of the source imagery that do not comprise the object region for the desired object can be joined with at least some of the one or more frames that do comprise the object region for the desired object. The resulting planar panorama imagery can thus comprise the desired object. Also, the planar panorama imagery can be generated so as to lack one or more non-desired objects and/or adverse effects that may result therefrom.

Planar panorama imagery generation

Among other things, one or more techniques and/or systems are disclosed for creating a three-dimensional model of an image. Image data, comprise a series of images of a location, such as along a route, can be received. The image data can comprise a first image element location for a first image element, where the first image element location may correspond to a three-dimensional point in the image, such as a location in the image with a depth from a point of observation. The first image element can be segmented into a first fa&#xe7;ade plane using the first image element location. The first fa&#xe7;ade plane can be merged with a second fa&#xe7;ade plane, resulting in a three-dimensional model of the image. The second fa&#xe7;ade plane can comprise a second image element, where the image data comprises a second image element location for the second image element.

Image three-dimensional (3D) modeling

Among other things, one or more techniques and/or systems are disclosed for selecting source imagery for a planar panorama comprising a curved path. The curved path can be identified in a block of source imagery, where the source imagery comprises inside-curve source imagery for an inside curve of the curved path. Image data from a first source image from the inside-curve source imagery can be compared with image data from a second source image from the inside-curve source imagery. If image data of the first source image sufficiently overlaps image data of the second source image the first source image can selected for inclusion in a set of images to be used for the planar panorama, while the second source image may not be so selected (e.g., to mitigate artifacts, object overlap, repeated objects, etc. in the planar panorama).

Source imagery selection for planar panorama comprising curve

Growing community of inventors

Bellevue, WA, United States of America

Don Dongwoo Kim