# US 12,602,876 B2 · Annotation tools for reconstructing three-dimensional roof geometry > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,602,876 B2 - Document ID: US12602876B2 - Status: Granted - Kind: B2 - Application number: 18/402,239 - Filed: 2024-01-02 - Published: 2024-08-29 - Granted: 2026-04-14 - Earliest claimed: 2023-02-23 - Assignees: Ecopia Tech Corporation - Applicants: Ecopia Tech Corporation - Inventors: Yuanming Shu; Shuo Tan; Dawei Zai - Web page: https://patentgrep.com/patent/US12602876B2 ## CPC - G06T 7/55: from multiple images - G06T 19/00: Manipulating three-dimensional [3D] models or images for computer graphics - G06F 30/12: characterised by design entry means specially adapted for CAD, e.g. graphical user interfaces [GUI] specially adapted for CAD - G06F 30/13: Architectural design, e.g. computer-aided architectural design [CAAD] related to design of buildings, bridges, landscapes, production plants or roads - G06T 7/62: of area, perimeter, diameter or volume - G06T 15/10: Geometric effects - G06T 19/20: Editing of three-dimensional [3D] images, e.g. changing shapes or colours, aligning objects or positioning parts - G06T 2200/24: involving graphical user interfaces [GUIs] - G06T 2207/10012: Stereo images - G06T 2207/20092: Interactive image processing based on input by user - G06T 2207/30184: Infrastructure - G06T 2210/04: Architectural design, interior design - G06T 2219/004: Annotating, labelling - G06T 2219/2021: Shape modification - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract Methods and systems for generating geometric models of buildings with complex roof geometry from multiview imagery are provided. An example method involves establishing a three-dimensional coordinate space for the multiview imagery, and displaying, through an annotation platform, first and second images of the multiview imagery depicting the pitched roof from different points of view. The method further involves receiving, through the annotation platform, user input annotating the first and second images with a plurality of roof elements defining a three-dimensional structure of the pitched roof, and reconstructing, based on the user input received through the annotation platform, a geometric model representing the three-dimensional structure of the pitched roof situated in the three-dimensional coordinate space. ## Claims ### Claim 1 A method for generating a geometric model of a pitched roof of a building from multiview imagery, wherein the method comprises: establishing a three-dimensional coordinate space for the multiview imagery; displaying, through an annotation platform, first and second images of the multiview imagery depicting the pitched roof from different points of view; providing, through the annotation platform, functionality for a user to provide user input comprising instructions to: (i) annotate a ridge line of the pitched roof as depicted in the first image, and (ii) adjust a height of the ridge line in the three-dimensional coordinate space to match a height of the ridge line as depicted in the second image; and reconstructing, based on the user input received through the annotation platform, a geometric model representing the three-dimensional structure of the pitched roof comprising the ridge line situated at the height of the ridge line in the three-dimensional coordinate space. ### Claim 2 (depends on 1) The method of claim 1, further comprising: providing, through the annotation platform, functionality for a user to provide user input comprising instructions to: define a perimeter surrounding a base of the pitched roof by: (a) annotating a perimeter of the pitched roof as depicted in the first image and (b) adjusting a height of the perimeter in the three-dimensional coordinate space to match a height of the perimeter as depicted in the second image. ### Claim 3 (depends on 2) The method of claim 2, further comprising: providing, through the annotation platform, functionality for a user to provide user input comprising instructions to: define connecting elements from first and second endpoints of the ridge line to the perimeter of the pitched roof. ### Claim 4 (depends on 1) The method of claim 1, further comprising: providing, through the annotation platform, functionality for a user to provide user input comprising instructions to define, during annotation of the ridge line, that the ridge line is to be parallel to a reference linear roof element. ### Claim 5 (depends on 1) The method of claim 1, further comprising: providing, through the annotation platform, functionality for a user to provide user input comprising instructions to annotate the ridge line as originating from a face of a planar roof facet. ### Claim 6 (depends on 1) The method of claim 1, wherein the first and second images comprise a stereoscopic pair. ### Claim 7 (depends on 1) The method of claim 1, wherein the first and second images each comprise an oblique image. ### Claim 8 (depends on 1) The method of claim 1, wherein the method further comprises: prior to receiving the user input, displaying a building identification guide in at least one of the first and second images to assist the user of the annotation platform to locate the building for annotation. ### Claim 9 (depends on 8) The method of claim 8, wherein the building identification guide comprises a two-dimensional polygonal representation of the building that is georeferenced to the three-dimensional coordinate space and projected into at least one of the first and second images. ### Claim 10 (depends on 9) The method of claim 9, wherein the two-dimensional polygonal representation was extracted from a single image comprising the first image or the second image. ### Claim 11 A system for generating a geometric model of a pitched roof of a building from multiview imagery, the system comprising one or more computing devices configured to: establish a three-dimensional coordinate space for the multiview imagery; display, through an annotation platform, first and second images of the multiview imagery depicting the pitched roof from different points of view; provide, through the annotation platform, functionality for a user to provide user input comprising instructions to: (i) annotate a ridge line of the pitched roof as depicted in the first image, and (ii) adjust a height of the ridge line in the three-dimensional coordinate space to match a height of the ridge line as depicted in the second image; and reconstruct, based on the user input received through the annotation platform, a geometric model representing the three-dimensional structure of the pitched roof comprising the ridge line situated at the height of the ridge line in the three-dimensional coordinate space. ### Claim 12 A non-transitory machine-readable storage medium comprising instructions that when executed cause one or more processors to: establish a three-dimensional coordinate space for multiview imagery; display, through an annotation platform, first and second images of the multiview imagery depicting a pitched roof from different points of view; provide, through the annotation platform, functionality for a user to provide user input comprising instructions to: (i) annotate a ridge line of the pitched roof as depicted in the first image, and (ii) adjust a height of the ridge line in the three-dimensional coordinate space to match a height of the ridge line as depicted in the second image; and reconstruct, based on the user input received through the annotation platform, a geometric model representing the three-dimensional structure of the pitched roof comprising the ridge line situated at the height of the ridge line in the three-dimensional coordinate space.