Xactware Solutions Inc v. Eagle View Technologies Inc

1. Case Identification

• Case #: IPR2016-00591
• Patent #: 8,209,152
• Filed: February 8, 2016
• Petitioner(s): Xactware Solutions, Inc.
• Patent Owner(s): Eagle View Technologies, Inc.
• Challenged Claims: 1, 3, 6, 8-10, 12-16, 19, 22-23, and 25

2. Patent Overview

• Title: Concurrent Display Systems and Methods for Aerial Roof Estimation
• Brief Description: The ’152 patent describes systems and methods for determining roof measurement information from multiple aerial images of a building. The claimed methods involve displaying at least two aerial images with different views of a roof, receiving a user indication of a roof feature in one image, modifying an underlying three-dimensional model based on that input, and then concurrently displaying a projection of the modified feature as a line drawing overlaid on corresponding locations of both aerial images to assist in verification and refinement.

3. Grounds for Unpatentability

Ground 1: Anticipation over Hsieh - Claims 1, 3, 6, 8-10, 12-16, 19, 22-23, and 25 are anticipated under 35 U.S.C. § 102(b) by Hsieh.

• Prior Art Relied Upon: Hsieh (a 1995 Carnegie Mellon publication entitled “Design and Evaluation of a Semi-Automated Site Modeling System”).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Hsieh’s semi-automated "SiteCity" system for extracting 3D building models from multiple images discloses every limitation of the challenged claims. Hsieh’s system displays multiple aerial images of a building in different "image windows," receives user input to "measure" a roof by outlining its features in one image, automatically generates a full 3D model of the building, and then projects that 3D model back onto the other aerial images as a 2D "line drawing" for verification and interactive modification. Petitioner asserted this process directly maps to the claims’ steps of displaying images, receiving an indication, modifying a model based on user corrections, and displaying an overlaid projection of features like ridgelines and edges.

Ground 2: Anticipation over Avrahami - Claims 1, 3, 6, 9-10, 12-16, 19, 22-23, and 25 are anticipated under 35 U.S.C. § 102(b) by Avrahami.

• Prior Art Relied Upon: Avrahami (a 2005 publication entitled “Extraction of 3D Spatial Polygons Based on the Overlapping Criterion for Roof Extraction from Aerial Images”).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner contended that Avrahami’s semi-automatic system for extracting 3D roof models from a pair of colored aerial images teaches all claimed elements. The system requires a user to manually point to a "seed point" on a roof plane in a first image. The system then automatically segments the roof polygon in that image, calculates an estimated height, transfers the seed point to a second image to segment the corresponding polygon, and performs an iterative matching process by "sliding" one polygon over the other to find maximum overlap, which generates a final 3D roof model. Petitioner argued that this process, including the display of overlaid polygons during the iterative matching, meets the limitations of the independent claims.

Ground 3: Obviousness over Hsieh and Perlant - Claim 10 is obvious over Hsieh in view of Perlant.

• Prior Art Relied Upon: Hsieh (1995 publication) and Perlant (a 1990 publication entitled “Scene Registration in Aerial Image Analysis”).

• Core Argument for this Ground:

  • Prior Art Mapping: Petitioner argued that Hsieh discloses all limitations of base claim 1, while Perlant supplies the additional limitations of dependent claim 10. Specifically, Perlant teaches a system where a user employs a marker to specify points on an aerial image, and based on these points, the system registers the aerial images to a reference grid corresponding to a 3D model. This registration process is precisely what claim 10 adds to claim 1.
  • Motivation to Combine: A POSITA would combine Perlant's registration functionality with Hsieh's 3D modeling system as a known technique to improve a similar device. Perlant explicitly suggests that scene registration is a required task for "the construction of a three-dimensional image," providing a direct motivation to integrate its teachings into a system like Hsieh to enhance the accuracy of the initial 3D model generation.
  • Expectation of Success: Incorporating a known registration technique (Perlant) into a known 3D modeling system (Hsieh) was a straightforward combination of prior art elements according to known methods that would predictably result in an improved system.

• Additional Grounds: Petitioner asserted numerous other obviousness grounds, including combinations of Hsieh with Applicad (to add roof estimate reporting) and Aerowest (to add network accessibility). A parallel set of obviousness grounds was asserted based on Avrahami in combination with Perlant, Applicad, and Aerowest.

4. Key Claim Construction Positions

• "reference grid": Petitioner proposed that this term be construed as "a data structure to which one or more images are registered." This construction was argued to be consistent with the patent’s specification and the understanding of a POSITA. It was critical for the obviousness argument against claim 10, as it framed the registration step as a well-understood concept, allowing Petitioner to map the functionality of the Perlant reference, which explicitly teaches registering images to a grid, onto the claim language.
• "roof estimate report": Petitioner proposed the construction "a representation of a roof which allows measurement information of at least one feature of the roof to be conveyed." This broad construction was important for the obviousness argument against claim 25, as it encompassed various forms of output beyond a simple list of measurements. It allowed Petitioner to argue that the detailed quotes, annotated drawings with dimensions, and material calculations generated by prior art software like Applicad constituted a "roof estimate report" as claimed.

5. Relief Requested

• Petitioner requests institution of an inter partes review of claims 1, 3, 6, 8-10, 12-16, 19, 22-23, and 25 of Patent 8,209,152 and cancellation of these claims as unpatentable.