Abstract: Embodiments relate to constructing a three-dimensional model using a plurality of two-dimensional photographic images. In a first embodiment, a first user input specifying a first position on a first photographic image may be received. A second user input specifying a second position on a second photographic image may be received, the second photographic image having a second perspective, specified by a second set of camera parameters, being different from a first perspective of the first image specified by a first set of camera parameters. The user inputs may indicate corresponding features in the images. Using photogrammetry, a point in a three-dimensional space may be determined according to the first position, the second position, the first set of camera parameters, and the second set of camera parameters. A surface of the three-dimensional model may be determined, whereby the surface may be curved to the point in three-dimensional space determined earlier.

Abstract: Embodiments relate to selecting textures for a panoramic photographic image in image-based three-dimensional modeling. In a first embodiment, a computer-implemented method includes a method for inputting a panoramic photographic image that uses a photogrammetry algorithm to adjust a plurality of camera parameters for the panoramic photographic image. In the method, a street-level panoramic photographic image is selected by a user from a street-level panoramic image display site. An area of the street-level panoramic image is also identified by the user. A photographic image at the area of the street-level panoramic image is captured. Camera parameters representing a camera that took the photographic image is determined. A position on the photographic image is mapped to a position on a three-dimensional model. The camera parameters are adjusted so that the three-dimensional model is rendered onto the photographic image from a perspective specified by the adjusted camera parameters.

Abstract: Embodiments relate to modeling three-dimensional interiors from photographic images. In a first embodiment, a computer-implemented method creates a three-dimensional model of an interior of a building from a photographic image. Input data is received from a user specifying a floor plan of the interior of the building. A first constraint input by a user indicating that a position on the floor plan corresponds to a position on the three-dimensional model is received. A second constraint input by a user indicating that a position on a photographic image of the interior corresponds to a position on the three-dimensional model is received. Finally, the three-dimensional model and camera parameters representing a camera that took the photographic image of the interior are altered based on the first and second constraints.

Abstract: Embodiments relate to locking geometric and camera parameters in image-based three-dimensional modeling. In a first embodiment, a computer-implemented method modifies a three-dimensional model using a set of two-dimensional photographic images. In the method, at least one constraint, input by a user, for a two-dimensional photographic image from the set of two-dimensional photographic images are received. Each constraint indicates that a position on the two-dimensional photographic image corresponds to a position on the three-dimensional model. When the number of constraints received for the two-dimensional photographic image is less than or equal to a first threshold value, camera parameters representing a camera that took the constrained photographic image. Finally, a photogrammetry algorithm is applied to at least the set of parameters to determine at least one unlocked parameter in the set of parameters and to modify the three-dimensional model based on the constraints.

Abstract: Embodiments relate to selecting and verifying textures in image-based three-dimensional modeling. In a first embodiment, a computer-implemented method for texturing a three-dimensional model. In the method, a polygon orientation of a polygon defining a face of a three-dimensional model is determined. A first camera orientation corresponding to a camera that took a first photographic image in a set of photographic images is also determined. The first photographic image is unconstrained to the three-dimensional model. The polygon orientation is compared with the first camera orientation to determine whether the first photographic image is directed toward the face of the three-dimensional model. If the first photographic image is directed toward the face of the three-dimensional model, a notification is provided for display to a user indicating that the first photographic image could be texture mapped to the three-dimensional model if the first photographic image is constrained to the three-dimensional model.

Abstract: A computer implemented system and method are disclosed for segmenting aerial and satellite imagery according to pre-defined regions contained in the images such as city blocks. An example system comprises a feature mask generation module, a partition generation module, one or more color map generation modules, a color map reassignment module, and a bounding box generation module. The system and method take aerial image data and vector map data as input and generate a composite color map for a geographic region of interest, wherein the color map comprises a plurality of pre-defined regions in which each pre-defined region can be defined by enclosing road segments. The disclosed system and method provide a convenient and logical way of segmenting large data sets of aerial imagery for use in other applications such as the construction of computer models of surface features of the Earth.

Abstract: In an embodiment, a method merges a plurality of three-dimensional models, each having been generated from images of a camera having a different viewpoint. For respective voxels in a plurality of voxels, a plurality of distance values are determined. Each determined distance value is a distance from the voxel to a three dimensional model from the plurality of three-dimensional models along a perspective of a camera model used to generate the three-dimensional model. When at least one of the determined distance values indicates that the distance between the voxel and the three dimensional model exceeds a threshold and the voxel is located above the corresponding three-dimensional model, a confidence score is determined indicating a degree to which the perspective of the camera model of the corresponding three-dimensional model is oriented to face the voxel. Based on the determined confidence score, the voxel into a merged three-dimensional model is determined.

Abstract: In an embodiment, a method determines a three-dimensional model from a plurality of images taken of a geographic region by one or more cameras from different perspectives. The method includes determining, using a first stereo reconstruction technique: (i) a plurality of three-dimensional candidate surface points from the plurality of images, and (ii) which of the images in the plurality of images view each of the plurality of candidate surface points. The method also includes identifying an empty space between each of the plurality of candidate surface points and each camera model for the respective images determined to view the candidate surface point. The method further includes for each of a plurality of pairs of images from the plurality of images, determining, using a second stereo reconstruction technique, a surface estimate for the pair of images. The method also includes merging the surface estimates to identify a final surface.

Abstract: A method of visualizing images of a target in a 3D modeling system is provided, according to an embodiment. The method may include mapping and displaying one or more images, generated from respective viewpoints of the target, on a visualization frame that encompasses a representation of the target based upon respective viewpoints of the images. In another embodiment, the method may include determining relevant pixels in the images and cropping the images based upon the determination of relevant pixels. According to a further embodiment, the method may include enabling a user to control the orientation of the visualization frame to display the images of the target corresponding with a specific viewpoint. A system for visualizing images of a target in a 3D modeling system may include a mapping module, a frame controller, a cropping module and interface storage, according to another embodiment.

Abstract: Embodiments relate to selecting and verifying textures in image-based three-dimensional modeling. In a first embodiment, a computer-implemented method for texturing a three-dimensional model. In the method, a polygon orientation of a polygon defining a face of a three-dimensional model is determined. A first camera orientation corresponding to a camera that took a first photographic image in a set of photographic images is also determined. The first photographic image is unconstrained to the three-dimensional model. The polygon orientation is compared with the first camera orientation to determine whether the first photographic image is directed toward the face of the three-dimensional model. If the first photographic image is directed toward the face of the three-dimensional model, a notification is provided for display to a user indicating that the first photographic image could be texture mapped to the three-dimensional model if the first photographic image is constrained to the three-dimensional model.

Abstract: Embodiments relate to locking geometric and camera parameters in image-based three-dimensional modeling. In a first embodiment, a computer-implemented method modifies a three-dimensional model using a set of two-dimensional photographic images. In the method, at least one constraint, input by a user, for a two-dimensional photographic image from the set of two-dimensional photographic images are received. Each constraint indicates that a position on the two-dimensional photographic image corresponds to a position on the three-dimensional model. When the number of constraints received for the two-dimensional photographic image is less than or equal to a first threshold value, camera parameters representing a camera that took the constrained photographic image. Finally, a photogrammetry algorithm is applied to at least the set of parameters to determine at least one unlocked parameter in the set of parameters and to modify the three-dimensional model based on the constraints.

Abstract: Embodiments relate to locking geometric and camera parameters in image-based three-dimensional modeling. In a first embodiment, a computer-implemented method modifies a three-dimensional model using a set of two-dimensional photographic images. In the method, at least one constraint, input by a user, for a two-dimensional photographic image from the set of two-dimensional photographic images are received. Each constraint indicates that a position on the two-dimensional photographic image corresponds to a position on the three-dimensional model. When the number of constraints received for the two-dimensional photographic image is less than or equal to a first threshold value, camera parameters representing a camera that took the constrained photographic image. Finally, a photogrammetry algorithm is applied to at least the set of parameters to determine at least one unlocked parameter in the set of parameters and to modify the three-dimensional model based on the constraints.

Abstract: A computer-assisted technique for constructing a three-dimensional model on top of one or more images (e.g., photographs) such that the model's parameters automatically match those of the real world object depicted in the photograph(s). Camera parameters such as focal length, position, and orientation in space may be determined from the images such that the projection of a three-dimensional model through the calculated camera parameters matches the projection of the real world object through the camera onto the image surface. Modeling is accomplished using primitives, such as boxes or pyramids, which may be intuitively manipulated to construct the three-dimensional model on a video display or other display screen of a computer system with a two-dimensional input controller (e.g., a mouse, joystick, etc.) such that the displayed three-dimensional object manipulation emulates physical three-dimensional object manipulation.

Abstract: A computer-assisted technique for constructing a three-dimensional model on top of one or more images (e.g., photographs) such that the model's parameters automatically match those of the real world object depicted in the photograph(s). Camera parameters such as focal length, position, and orientation in space may be determined from the images such that the projection of a three-dimensional model through the calculated camera parameters matches the projection of the real world object through the camera onto the image surface. Modeling is accomplished using primitives, such as boxes or pyramids, which may be intuitively manipulated to construct the three-dimensional model on a video display or other display screen of a computer system with a two-dimensional input controller (e.g., a mouse, joystick, etc.) such that the displayed three-dimensional object manipulation emulates physical three-dimensional object manipulation.

Abstract: A computer-assisted technique for providing approximate solutions to photogrammetric problems in interactive applications computes an approximate solution to a large, non-linear system by converting the non-linear system into a smaller, linear system where only a subset of the system's parameters are used. The size of the smaller system can be adjusted to different system capabilities and/or model complexities to ensure computation of an approximate solution in a given time. For use in an interactive environment, this time may be adjusted so that redraw rates are within acceptable tolerances, say from 5 to 30 frames per second. The method may allow for drawing a three-dimensional representation of a scene using parameterized primitives having constraints which are updated interactively, so as to minimize error in constrained locations of the primitives. Such drawing may include creating a Jacobian matrix for a predetermined minimum set of parameters of the primitives.

Abstract: A computer-assisted technique for constructing a three-dimensional model on top of one or more images (e.g., photographs) such that the model's parameters automatically match those of the real world object depicted in the photograph(s). Camera parameters such as focal length, position, and orientation in space may be determined from the images such that the projection of a three-dimensional model through the calculated camera parameters matches the projection of the real world object through the camera onto the image surface. Modeling is accomplished using primitives, such as boxes or pyramids, which may be intuitively manipulated to construct the three-dimensional model on a video display or other display screen of a computer system with a two-dimensional input controller (e.g., a mouse, joystick, etc.) such that the, displayed three-dimensional object manipulation emulates physical three-dimensional object manipulation.

Abstract: Information extracted from two or more images of a scene is composited into a single texture image for each surface of a computer-generated model of the scene thus providing, view-independent rectangular textures over standard geometries. In conjunction with a photogrammetric modeling system, the texture information may be extracted from a variety of images, which may be taken from a variety of different angles, with different cameras and under different lighting conditions. The fusion of the texture information is preferably performed during a preprocessing step, eliminating the need for further texture processing. Perspective projection artifacts in the final texture is reduced or eliminated. Further, visible seams between partial textures from different images (which may otherwise be caused by misregistration or differences in illumination) are reduced or avoided.

Abstract: A method of testing a computer program in a computer system is described. The computer system includes a source code corresponding to the computer program, and a number of tests. The method includes the following steps. A coverage point is inserted into the source code to correspond to a statement in the source code. An executable, corresponding to the source code, is executed using the tests. This generates a result. The result is used to generate a subset of tests that executed the coverage point and tested the statement. Test a second executable using the subset of tests. The second executable corresponds to the source code.