PTAB

IPR2021-00937

Matterport Inc v. Appliance Computing III Inc

Key Events
Petition
petition

1. Case Identification

2. Patent Overview

  • Title: Image-Based Rendering of Real Spaces
  • Brief Description: The ’111 patent describes systems and methods for rendering three-dimensional (3D) models of real estate spaces. The technology involves capturing images of a property from multiple viewpoints, processing the image data to generate a virtual model, and displaying the model through a user interface that facilitates virtual tours with navigation tools.

3. Grounds for Unpatentability

Ground 1: Obviousness of Claims 1-23 over Cowtan in view of Bell436

  • Prior Art Relied Upon: Cowtan (Application # 2009/0031246) and Bell436 (Application # 2014/0043436).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that the combination of Cowtan and Bell436 teaches every limitation of the challenged claims. Cowtan disclosed an internet-based, dual-paned virtual tour system for the real estate industry, providing a framework for displaying property images alongside interactive 2D floor plans and navigation tabs (e.g., "First Floor," "Second Floor"). However, Cowtan's system was based on simpler panoramic images. Bell436, which describes Petitioner's own prior art technology, taught advanced methods for capturing and reconstructing detailed 3D scenes from images taken by devices like smartphones. Bell436 specifically disclosed generating 3D models, point clouds, and textured meshes from panoramic images, and defining spatial boundaries (e.g., floors, walls) from the resulting 3D data. Petitioner contended that implementing Bell436's sophisticated 3D modeling techniques within Cowtan's established real estate tour framework would result in the claimed invention. For instance, Cowtan taught receiving images corresponding to rooms, and Bell436 taught creating panoramas and 3D models from such images, together meeting the core limitations of independent claims 1 and 18.
    • Motivation to Combine: Petitioner asserted a person of ordinary skill in the art (POSITA) would combine these references to improve upon the known virtual tour systems disclosed in Cowtan. A POSITA would have looked to Bell436's well-documented methods for creating more realistic and immersive 3D virtual models to enhance the user experience of a real estate tour platform like Cowtan's. Petitioner noted that the ’111 patent’s provisional application acknowledged Bell436’s underlying technology (Matterport’s) as a pre-existing commercial example of image-based rendering.
    • Expectation of Success: The combination involved applying known 3D reconstruction techniques (Bell436) to a known application (Cowtan's real estate tours) to achieve the predictable result of a more advanced virtual tour. A POSITA would have had a reasonable expectation of successfully integrating these technologies.

Ground 2: Obviousness of Claim 9 over Cowtan, Bell436, and Triggs

  • Prior Art Relied Upon: Cowtan (Application # 2009/0031246), Bell436 (Application # 2014/0043436), and Triggs (a 2000 journal article titled "Bundle Adjustment — A Modern Synthesis").
  • Core Argument for this Ground:
    • Prior Art Mapping: This ground built upon the combination of Cowtan and Bell436 to address the specific limitation of claim 9, which required grouping panoramas "according to a panorama bundle adjustment heuristic." Petitioner argued that while Bell436 taught alignment techniques for 3D captures generally, Triggs was a foundational reference that explicitly described various "bundle adjustment" techniques. Triggs taught that bundle adjustment is a method for "refining a visual reconstruction to produce jointly optimal 3D structure and viewing parameter" estimates, which directly corresponded to the heuristic required by claim 9.
    • Motivation to Combine: A POSITA, having combined Cowtan and Bell436 to create 3D virtual tours, would be motivated to improve the accuracy and quality of the 3D scene reconstruction. Bundle adjustment, as taught comprehensively by Triggs, was a well-known and powerful method for achieving such refinement in 3D computer vision. Therefore, it would have been a natural and obvious step to incorporate the bundle adjustment techniques from Triggs into the 3D modeling process of Bell436.
    • Expectation of Success: Given that bundle adjustment was a standard optimization technique in the field of 3D reconstruction by the priority date, a POSITA would have had a high expectation of success in applying it to the methods taught by Bell436.

4. Arguments Regarding Discretionary Denial

  • Petitioner presented substantial arguments that discretionary denial would be inappropriate under both 35 U.S.C. §325(d) and the Fintiv factors.
  • §325(d): Petitioner argued that although Cowtan and Bell436 were cited on an Information Disclosure Statement (IDS) during prosecution, the examiner never used them in any rejection or provided evidence that their teachings were substantively considered on the merits. Therefore, the arguments and combination presented in the petition were new.
  • Discretionary Denial under Fintiv: Petitioner argued the factors weighed heavily against denial. Key arguments included: (1) the parallel district court litigation was in its early stages, with a trial date that was likely to be rescheduled; (2) the invalidity grounds in the petition were distinct from those in the litigation; (3) the defendant in the litigation (Redfin Corp.) offered to stipulate that it would not pursue the same invalidity grounds if the inter partes review (IPR) was instituted; and (4) most critically, the Petitioner (Matterport) was not a party to the parallel litigation, and denying institution would deprive it of the efficient administrative process intended by Congress while it faced the threat of future infringement suits from the Patent Owner.

5. Relief Requested

  • Petitioner requested institution of an IPR and cancellation of claims 1-23 of the ’111 patent as unpatentable under 35 U.S.C. §103.