Image-based rendering of real spaces

  • US 9,836,885 B1
  • Filed: 10/27/2014
  • Issued: 12/05/2017
  • Est. Priority Date: 10/25/2013
  • Status: Active Grant
First Claim
Patent Images

1. A computer system for providing image-based rendering of real property, the computer system comprising:

  • a processor implementing program commands and processing data;

    a memory storing the program commands and the processing data, the memory being operatively coupled to the processor;

    a user interface visually presenting an image-based rendering of a real property, the user interface being operatively coupled with the memory and processor;

    wherein the program commands cause the processor to;

    obtain two or more photorealistic viewpoints from capture locations in a plurality of spaces of the real property using ground truth image data, the ground truth image data including images of the real property captured from the capture locations;

    combine and process two or more instances of ground truth image data to create a plurality of synthesized viewpoints;

    generate a virtual model of a current space from the plurality of spaces in the real property from the perspective of a current viewpoint, the current viewpoint being one of the plurality of photorealistic viewpoints or one of the plurality of synthesized viewpoints associated with the current space;

    identify the current space in the real property where the current viewpoint is located using metadata associated with the ground truth data;

    visually present the virtual model on a first portion of the user interface, a map overlay on a second portion of the user interface, and a text overlay on a third portion of the user interface, the map overlay indicating a position of the current viewpoint in the current space, the text overlay including a first label identifying the current space in the real property where the current viewpoint is located;

    generate a plurality of spatial boundaries of the plurality of spaces of the real property, each of the plurality of spatial boundaries defining a parcel outline of a corresponding space among the plurality of spaces; and

    generate the ground truth image data by annotating the images of the real property with second labels indicating the capture locations using the plurality of spatial boundaries,wherein the virtual model, the map overlay, and the text overlay are functionally linked based on the position of the current viewpoint in the real property, andwherein identifying the current space in the real property where the current viewpoint is located using the metadata associated with the ground truth data includes identifying the first label of the current space using the second labels indicating the capture locations.

View all claims
    ×
    ×

    Thank you for your feedback

    ×
    ×