PTAB

IPR2025-01358

Zestyai Inc v. Aon Re Inc

Log In
Key Events
Petition
petition

1. Case Identification

2. Patent Overview

  • Title: Property Characteristic Classification and Condition Analysis
  • Brief Description: The ’029 patent discloses methods and systems for automatically analyzing aerial imagery to evaluate property characteristics (e.g., roof type), classify their condition, and estimate the risk of damage to the property from potential disasters. The system uses machine learning classifiers to extract features from images and provide a condition assessment to a user.

3. Grounds for Unpatentability

Ground 1: Obviousness over Gross and Davis - Claims 1-7 and 15-20 are obvious over Gross in view of Davis.

  • Prior Art Relied Upon: Gross (Application # 2015/0186953) and Davis (Patent 10,755,357).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that Gross teaches nearly all limitations of the independent claims. Gross discloses a property assessment system using a "Classifier Engine" (a machine learning classifier) to analyze images of properties, identify structural elements like roofs ("property characteristics"), and determine their type and condition (e.g., "damaged"). The system is trained on reference images to automatically perform these classifications. Petitioner asserted that Davis supplies the specific teaching of estimating risk from "one or more disasters." While Gross mentions assessing risk from hazards like "arson," Davis explicitly discloses analyzing aerial images to determine risks from natural disasters like fires and hurricanes, based on property features.
    • Motivation to Combine: A Person of Ordinary Skill in the Art (POSA) would combine Davis's explicit disaster-based risk assessment with Gross's more general property condition analysis system. The combination would provide insurers with more comprehensive and useful risk information, a clear benefit. Petitioner contended this would have been a predictable variation, as Gross already contemplated developing "correlations" for insurance risk, and Davis provided a known method for creating such correlations for specific disasters.
    • Expectation of Success: A POSA would have a reasonable expectation of success because both Gross and Davis employ similar, well-known image analysis and machine learning techniques to assess property features for insurance risk purposes. Integrating Davis's disaster-specific rules into Gross's framework would be a straightforward application of known data correlation methods.

Ground 2: Obviousness over Gross - Claims 8 and 11-14 are obvious over Gross.

  • Prior Art Relied Upon: Gross (Application # 2015/0186953).
  • Core Argument for this Ground:
    • Prior Art Mapping: This ground targets system claims 8 and 11-14. Petitioner argued that Gross's disclosure of a system that performs the challenged method renders the corresponding system claims obvious. Gross describes its processes being implemented via "executable software routines and modules" on a computing system with "processing circuitry," which directly maps to the system limitations of claim 8. For dependent claims, Petitioner argued Gross teaches determining a "replacement cost" by disclosing the calculation of "repair or improvement figures" (claim 8), comparing current and historic property conditions to identify "structural improvements" over time (claims 11 and 12), and assessing image "orthogonality" through its incorporation of Pershing (Patent 8,078,436) (claims 13 and 14).
    • Motivation to Combine: Not applicable, as this ground relies on a single reference. Petitioner asserted that all claimed features are either explicitly taught or would have been obvious modifications of Gross's disclosed system.

Ground 3: Obviousness over Gross and Furukawa - Claims 9 and 10 are obvious over Gross in view of Furukawa.

  • Prior Art Relied Upon: Gross (Application # 2015/0186953) and Furukawa (Patent 6,970,593).
  • Core Argument for this Ground:
    • Prior Art Mapping: This ground challenges dependent claims 9 and 10, which add limitations requiring the system to obtain a "shape map image," overlay it with the aerial image, and determine if property boundaries match. Petitioner argued Gross identifies a need for confirming tentative property addresses tagged during its analysis. Furukawa directly addresses this problem by teaching a system that obtains map data containing the "external shape of buildings" and compares it to shapes extracted from satellite images to confirm property locations.
    • Motivation to Combine: A POSA would be motivated to incorporate Furukawa's map-based confirmation method into Gross's system to solve the address confirmation problem explicitly identified in Gross. Furukawa provides a known, readily available solution to a problem inherent in Gross's system. The combination would improve the accuracy of Gross's property assessment by ensuring the system is analyzing the correct property.
    • Expectation of Success: A POSA would have a reasonable expectation of success in combining the references, as it involves applying Furukawa's established technique of overlaying map data onto the aerial imagery already being used by Gross.

4. Relief Requested

  • Petitioner requests the institution of an inter partes review (IPR) and the cancellation of claims 1-20 of Patent 10,529,029 as unpatentable.