Systems and methods for generating three dimensional geometry
Systems and methods are described for creating three dimensional models of building objects by creating a point cloud from a plurality of input images, defining edges of the building object's surfaces represented by the point cloud, creating simplified geometries of the building object's surfaces and constructing a building model based on the simplified geometries. Input images may include ground, orthographic, or oblique images. The resultant model may be scaled according to correlation with select image types and textured.
1. One or more non-transitory computer readable medium storing instructions that, when executed, cause a processor to execute operations comprising:
receiving a plurality of ground-level images of an entire single building captured by a capture device circumnavigating the entire single building;
constructing a textured contiguous multi-dimensional building model of the entire single building by processing external surfaces of the entire single building identified within the plurality of ground-level images, wherein the constructing includes at least defining boundaries for a plurality of planar surfaces of the entire single building by recognition of building vertices, correlating and aligning the plurality of planar surfaces to one another along the defined boundaries to form a contiguous multi-dimensional model, and re-texturing the plurality of planar surfaces with imagery from the plurality of ground-level images of the entire single building, and wherein the external surfaces include one or more architectural features;
scaling the imagery re-textured contiguous multi-dimensional building model using a scaling factor based on a comparison of dimensional ratios of the one or more architectural features as identified in the plurality of captured ground-level images, to known dimensional ratios of industry standard architectural features, wherein the scaling factor is based on dimensional ratios of the one or more architectural features as identified in the plurality of captured ground-level images; and
storing the scaled imagery re-textured contiguous multi-dimensional building model.
2. The one or more non-transitory computer readable medium of claim 1 , wherein the one or more architectural features include roofing surfaces or roof intersection vertices.
3. The one or more non-transitory computer readable medium of claim 1 , wherein the one or more architectural features include one or more of: ground intersection vertices, bay window vertices, or overhanging vertices.
4. The one or more non-transitory computer readable medium of claim 1 , further storing instructions that, when executed, cause the processor to execute operations comprising identifying the one or more architectural features by matching to stored libraries of known architectural features.
5. A method for generating a multi-dimensional building model, the method comprising:
receiving a plurality of ground-level images of an entire single building captured from multiple angles and distances around the entire single building;
creating a textured contiguous multi-dimensional building model by constructing a plurality of facades of the entire single building based on the plurality of ground-level images, the plurality of facades including all sides and roof of the entire single building, wherein the constructing includes at least defining boundaries for a plurality of facades of the entire single building by recognition of building vertices, correlating and aligning the facades to one another along the defined boundaries to form a contiguous multi-dimensional model, and re-texturing the facades with imagery from the plurality of ground-level images of the entire single building;
determining scale of the imagery re-textured contiguous multi-dimensional building model using a scaling factor based on comparing dimensional ratios of one or more building architectural features located within one or more of the plurality of facades to known dimensional ratios of industry standard architectural features;
reconstructing the imagery re-textured contiguous multi-dimensional building model from the determined scale; and
storing the reconstructed imagery re-textured contiguous multi-dimensional building model.
6. One or more non-transitory computer readable medium storing instructions that, when executed, cause a processor to execute operations comprising:
receiving a plurality of ground-level images of an entire single building captured from multiple angles and distances around the entire single building;
creating a textured contiguous multi-dimensional building model by constructing a plurality of facades of the entire single building based on the plurality of ground-level images, the plurality of facades including all sides and roof of the entire single building, wherein the constructing includes at least defining boundaries for a plurality of facades of the entire single building by recognition of building vertices, correlating and aligning the facades to one another along the defined boundaries to form a contiguous multi-dimensional model, and re-texturing the facades with imagery from the plurality of ground-level images of the entire single building;
determining scale of the imagery re-textured contiguous multi-dimensional building model using a scaling factor based on comparing dimensional ratios of one or more building architectural features located within one or more of the plurality of facades to known dimensional ratios of industry standard architectural features;
reconstructing the imagery re-textured contiguous multi-dimensional building model from the determined scale; and
storing the reconstructed imagery re-textured contiguous multi-dimensional building model.