Abstract: Computing a height of a building is performed by inputting a pair of two-dimensional (2-D) aerial images of a city along with their metadata. Using the metadata, a three-dimensional (3-D) vector from each image toward the location of the camera when each image was taken is determined. A plurality of pairs of corresponding image points from the images are computed, in each pair of image points an image point of one image identifies the same physical point on the building as the second image point of the second image. Next, the images are superimposed, and for each pair of image points, determine the intersection of the 3-D vector of the first image originating at the first image point with the 3-D vector of the second image originating at the second image point. Each intersection is a 3-D position and the height is determined from the median of these 3-D positions.

Abstract: Vectorization of an image begins by receiving a two-dimensional rasterized image and returning a descriptor for each pixel in the image. Corner detection returns coordinates for all corners in the image. The descriptors are filtered using the corner positions to produce corner descriptors for the corner positions. A score matrix is extracted using the corner descriptors in order to produce a permutation matrix that indicates the connections between all of the corner positions. The corner coordinates and the permutation matrix are used to perform vector extraction to produce a machine-readable vector file that represents the two-dimensional image. Optionally, the corner descriptors may be refined before score extraction and the corner coordinates may be refined before vector extraction. A three-dimensional or N-dimensional image may also be input.

Abstract: Two-dimensional objects are displayed upon a user interface; user input selects an area and selects a machine learning model for execution. The results are displayed as an overlay over the objects in the user interface. User input selects a second model for execution; the result of this execution is displayed as a second overlay over the objects. A first overlay from a model is displayed over a set of objects in a user interface and a ground truth corresponding to the objects is displayed as a second overlay on the user interface. User input selects the ground truth overlay as a reference and causes a comparison of the first overlay with the ground truth overlay; the visual data from the comparison is displayed on the user interface. A comparison of M inference overlays with N reference overlays is performed and visual data from the comparison is displayed on the interface.

Abstract: Input coordinates identify a tile on the globe and a biome is identified for each pixel of the tile. We choose a random segmentation mask of the corresponding biome for the tile. Every segment of the mask carries a unique identifier and access to a random value via a corner reference. These parameters ensure that (when executed in parallel) each pixel with the same identifier and random value is handled the same way. For each pixel of the tile, using the selected mask, we identify the segment corresponding to that pixel, we retrieve a color variation for that pixel based upon the segment identifier and the random value. We thus choose a color variation for each segment of a segmentation mask. Pixels in the same segment are treated the same way. Instead of a color variation for flat areas (crop fields) we choose a density variation for regions like forests.

Abstract: Two-dimensional aerial images and other geo-spatial information are processed to produce land classification data, vector data and attribute data for buildings found within the images. This data is stored upon a server computer within shape files, and also stored are source code scripts describing how to reconstruct a type of building along with compiled versions of the scripts. A software game or simulator executes upon a client computer in which an avatar moves within a landscape. A classifier classifies a type of building in the shape file to execute the appropriate script. Depending upon its location, a scene composer downloads a shape file and a compiled script is executed in order to reconstruct any number of buildings in the vicinity of the avatar. The script produces a three-dimensional textured mesh which is then rendered upon a screen of the client computer to display a two-dimensional representation of the building.

Abstract: Workload scheduling allows geospatial dependencies between workload items, which may vary in size, grouping or composition from workflow step to workflow step and can automatically adjoin existing data. By allowing dependencies between workload items in different steps, processing (and finish processing) of a dependent workload item begins in a subsequent step once its required workload items in a previous step have finished, thus enabling the processing of that dependent workload item before the previous step has completely processed all of its workload items. A dependency may be created because an item in a subsequent step intersects spatially with an item from a previous step, or because within a current step items have a buffer, necessitating that one item may depend upon spatially neighboring items. Workloads in two dimensions, three dimensions and higher may be used, and may be grid based, general polygons, or other in shape.

Abstract: An interactive learning cycle includes an operator, a computer and a pool of images. The operator produces a sparsely-labeled data set. A back-end system produces live feedback: a densely-labeled training set which is displayed on the computer. Immediate feedback is displayed in color on the operator computer in less than about five seconds. A labeling tool displays a user interface and for every labeling project a region is defined that is downloaded as an image data batch. The operator annotates on a per-image basis in the region and uses several UI tools to mark features in the image and group them to a predefined label class. The back-end system includes processes that run in parallel and feed back into each other, each executing a model. A local model is used independently of the global model. The global model accepts sparsely-labeled images from numerous operator computers.